CERN: Climate Models will need to be revised

[TimG]

I just have to ignore this 'upside down' issue. It's not a word that is used in statistics, and I haven't yet understood what MM meant by their comments on that.

I explained it here:

i.e. the physics of the lake sediments says they 'increased' with temperature. The paper assumed the 'decreased' with temperatures. Mann claims that his algorithm does not care but that is a lie. No algorithm allows one to use proxies in an orientation that opposite to what is required by the underlying physics of the proxy. To put it another way: if Mann assumed that tree width decreases as temperature increases then he would have made the same mistake he made with the lake sediments.

What don't you understand?

[Michael Hardner]

What don't you understand?

As I said, it seems that he either misunderstood the correlation completely - which can't be the case - or he had a negative correlation factor in there in which case who cares?

[noahbody]

guys, guys... yours is quite a favourite for WTFIUWT TV-weatherman devotees!

alternatively, for your consideration and edification:

So the melting of iceburgs, disappearance of landmarks, change in fauna can all be chalked up as weather? If you didn't click on the article, it's a detailed account from a sea captain who sailed the arctic for 54 years. If anything it indicates the GISST data from that era is far from accurate.

[TimG]

As I said, it seems that he either misunderstood the correlation completely - which can't be the case - or he had a negative correlation factor in there in which case who cares?

Huh? The problem is the last 200 years of data is junk and happens to have a correlation which is opposite the correlation that the physics dictates. He ignored the warning labels on the data and ran it through his meat grinding algorithm which applied a negative correlation when it was supposed to be positive. IOW, he used the data upside down.

Given the evidence it is either: 1) incredibly sloppy work or 2) deliberate fraud on the part of Mann. No matter which it is the scientific community should care that he has refused to publish a correction to his work that makes it clear that he has not produced a reconstruction which does not depend on tree rings. You are a good example of why a correction is necessary since you were fooled into believing that he had produced a reconstruction which does not depend on tree rings.

Edited September 12, 2011 by TimG

[Michael Hardner]

Huh? The problem is the last 200 years of data is junk and happens to have a correlation which is opposite the correlation that the physics dictates. He ignored the warning labels on the data and ran it through his meat grinding algorithm which applied a negative correlation when it was supposed to be positive. IOW, he used the data upside down.

Given the evidence it is either: 1) incredibly sloppy work or 2) deliberate fraud on the part of Mann. No matter which it is the scientific community should care that he has refused to publish a correction to his work that makes it clear that he has not produced a reconstruction which does not depend on tree rings. You are a good example of why a correction is necessary since you were fooled into believing that he had produced a reconstruction which does not depend on tree rings.

He did produce a reconstruction, but you rejected it for almost the same reasons that you rejected tree ring data.

I'm still not clear on what happened... using the data upside-down still doesn't make sense to me. If he applied a negative to a negative and it came out positive anyway, then who cares ? It's just notation.

[TimG]

He did produce a reconstruction, but you rejected it for almost the same reasons that you rejected tree ring data.

It is not just me. The scientist that collected the data criticized Mann for misusing it.

I'm still not clear on what happened... using the data upside-down still doesn't make sense to me. If he applied a negative to a negative and it came out positive anyway, then who cares ? It's just notation.

It is clear I am wasting my time trying to explain these issues. You simply do not have the background required to understand the issues and seem to think that climate scientists are infallible. The short answer is Mann screwed up again. His reconstruction is junk and has no merit. Anyone who is not CAGW fanatic and has any knowledge of statistics recognizes this.

[Michael Hardner]

It is clear I am wasting my time trying to explain these issues. You simply do not have the background required to understand the issues and seem to think that climate scientists are infallible. The short answer is Mann screwed up again. His reconstruction is junk and has no merit. Anyone who is not CAGW fanatic and has any knowledge of statistics recognizes this.

Actually I took enough undergraduate statistics (Math major) so I understand how to do regression analysis and the like. I just don't remember ever hearing the term 'upside down' being used.

You should be able to add any factor to the analysis and see if it's significant, irregardless of sign.

It's not clear from the record that Mann screwed up. Your summary of it as 'junk' and having 'no merit' is basically what he said of the criticisms. So, we're left with waiting to hear back from MM I guess about his responses.

[TimG]

Actually I took enough undergraduate statistics (Math major) so I understand how to do regression analysis and the like. I just don't remember ever hearing the term 'upside down' being used.

As I said above. The data for the last 200 years is contaminated and should not be used. Mann used this contaminated data to determine the 'orientation' for the data and picked a direction that makes no physical sense. i.e. he used the data upside down. Why is this hard to understand? Do you understand that proxies have physical relationship to the real world and this relation cannot be changed by statistical algorithms?

Actually So, we're left with waiting to hear back from MM I guess about his responses.

PNAS does not print comments on comments. If you want to know MM's response you will have read ClimateAudit. Edited September 12, 2011 by TimG

[Michael Hardner]

As I said above. The data for the last 200 years is contaminated and should not be used. Mann used this contaminated data to determine the 'orientation' for the data and picked a direction that makes no physical sense. i.e. he used the data upside down. Why is this hard to understand? Do you understand that proxies have physical relationship to the real world and this relation cannot be changed by statistical algorithms?

Ok, how about an example.

Say I'm trying to correlate sea depths to air temperature directly above the ocean.

If I use the data (metres below sea level) as 5, 10, 50, 100 etc. and temperatures (in C) 20, 21, 22,23 etc. OR I use the data as (point distance from zero) as -5, -10, -50, -100 etc...

Well it shouldn't matter - if there's a correlation, then it will be the same number for both cases but one will be negative. This is what I'm stuck on.

There are no 'upside down' numbers in math, just negatives.

[TimG]

If I use the data (metres below sea level) as 5, 10, 50, 100 etc. and temperatures (in C) 20, 21, 22,23 etc. OR I use the data as (point distance from zero) as -5, -10, -50, -100 etc...

This has nothing to do with what Mann is doing.

Let's say you want to use the volume of mercury in a tube as a proxy for temperature where volume is measured mm.

Lets say you take for measurements 1mm, 2mm, 3mm and you correlated them with temperatures 2, 3 and 4 degC.

That data says an increase of 1mm equals 1 degC increase.

All good? Right?

What happens if you switch the sign. -1mm, -2mm, -3mm and do the correlation.

This data says a DECREASE of 1mm equals 1 degC.

But the volume of mercury decreasing as temperature increases makes no sense because mecury expands when it is heated!

Your example assumed that he simply shifted the series where 3mm -> -1 mm and 1mm -> -3mm.

Mann changed the sign so 3mm -> -3mm and 1mm > -1mm

It is different operation with different mathematical results.

He took a proxy where a increase in thickness means an increase in temperatures and treated it as if an DECREASE in thickness means an increase in temperatures. It is wrong. There is no interpretation to debate. It is flat out wrong.

Edited September 12, 2011 by TimG

[eyeball]

It is a little different. With the lake sediments the scientists have identified specific physical mechanisms (agricultural runoff into the lakes) that explains the divergence. This means the do have some confidence in the data prior to 200 years ago. The trouble is this information cannot be calibrated to thermometers - all it does is tell us the relative changes over time. Incidently - the lake sediments show a strong MWP warming period - an artifact that was suppressed when Mann used them upside down.

I bet $10 the scientists, bureaucrats and lobbyists working for the people directing their agricultural runoff into these ecosystems will tell you the runoff is completely harmless. $20 says their critics are a regarded like a bunch of hysterical communists too.

[Michael Hardner]

He took a proxy where a increase in thickness means an increase in temperatures and treated it as if an DECREASE in thickness means an increase in temperatures. It is wrong. There is no interpretation to debate. It is flat out wrong.

That's an incredible mistake.

[eyeball]

That's an incredible mistake.

What's really incredible is that some 95% or more of scientists are involved in a grand global conspiracy to overlook this for the purpose of ensuring their research funding.

If this many scientists from just one field of expertise are this corrupt what am I to assume or believe about other sciences, like economics for example?

Never mind who am I to believe these days exactly HOW am I supposed to believe them?

Edited September 12, 2011 by eyeball

[Sir Bandelot]

Lets say someone gives you 1 million dollars. Next, another person gives you ten dollars. How much do you have? Basically, a million dollars.

Let's say, instead of gives you ten dollars, they remove ten dollars. How much do you have now?

It could simply be a question of scale. If one component predominates by a factor of 100 or more, the sign of the other is irrelevant.

[TimG]

That's an incredible mistake.

As in you agree with me or you still don't believe me?

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.

2) Taking the lake sediment proxies out does not change the result.

Reply: that is because Mann also uses the tree proxies such as Bristle cone pines which are mostly likely not temperature proxies. If you take the lake sediment proxies and tree ring proxies out you end up with a reconstruction that has no significance prior to 1500. Mann admited this in the SI for Mann 2009.

3) Mann acknowledged this the SI of Mann 2009.

Reply: This problem was pointed in a comment submitted to PNAS by McIntryre. Therefore the only acceptable response for Mann to submit a correction to Mann 2008 that acknowledges the issue and withdraws the Mann 2008 claim that the reconstruction is valid without tree rings.

Edited September 13, 2011 by TimG

[TimG]

What's really incredible is that some 95% or more of scientists are involved in a grand global conspiracy to overlook this for the purpose of ensuring their research funding.

What is incredible is people who create irrelevant strawmen and pat themselves on the back when the demolish them.

Tell me: are 3 billion Christians involved in a grand conspiracy to convince the world that Jesus is the Son of God?

I find it incredibly unlikely that so many people could knowingly collude to believe in a lie. Therefore Jesus must be the Son of God.

If there are problems with climate science it is because climate scientists have convinced themselves that falsehoods are true. It is not because they are knowingly engaging in deception.

We know this happens all of the time in human society and it has even happened in the past with scientists (bacteria and ulcers are one example). Therefore, it is most certainly possible that 97% of the scientists are wrong about critical aspects of climate science such as climate sensitivity.

Edited September 13, 2011 by TimG

[TimG]

It could simply be a question of scale. If one component predominates by a factor of 100 or more, the sign of the other is irrelevant.

In this case the flipped sign was on the data that dominated the reconstruction so it is most definately relevant.

[eyeball]

What is incredible is people who create irrelevant strawmen and pat themselves on the back when the demolish them.

Who's strawman? I though the grand global conspiracy to swindle the public for AGW funding research was a matter of faith with folks like you?

Tell me: are 3 billion Christians involved in a grand conspiracy to convince the world that Jesus is the Son of God?

The evangelical one's certainly are and they're not ashamed of their faith like you suddenly seem to be.

[TimG]

Who's strawman? I though the grand global conspiracy to swindle the public for AGW funding research was a matter of faith with folks like you?

I have never said that. I think that many people on the left like climate change as an issue because it gives them justifications to pass policies that they believe are good no matter what. But saying that people are motivated by their beliefs is a lot different than saying they are out to swindle people.

The evangelical one's certainly are and they're not ashamed of their faith like you suddenly seem to be.

Who said I am an evangelical? Who said I am even a Christian? Edited September 13, 2011 by TimG

[Michael Hardner]

As in you agree with me or you still don't believe me?

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.

2) Taking the lake sediment proxies out does not change the result.

Reply: that is because Mann also uses the tree proxies such as Bristle cone pines which are mostly likely not temperature proxies. If you take the lake sediment proxies and tree ring proxies out you end up with a reconstruction that has no significance prior to 1500. Mann admited this in the SI for Mann 2009.

3) Mann acknowledged this the SI of Mann 2009.

Reply: This problem was pointed in a comment submitted to PNAS by McIntryre. Therefore the only acceptable response for Mann to submit a correction to Mann 2008 that acknowledges the issue and withdraws the Mann 2008 claim that the reconstruction is valid without tree rings.

I feel I have reached the edge of my ability to understand this. I don't have faith that you understand it either - you may just be quoting McIntyre here for all I know.

I just can't believe Mann could be that clueless. That would only be possible if there were a conspiracy.

[Bonam]

I feel I have reached the edge of my ability to understand this.

Precisely why it is pointless to debate technical scientific issues on forums such as these. It is dubious at best even on forums specifically dedicated to the debate of technical scientific issues.

I don't have faith that you understand it either - you may just be quoting McIntyre here for all I know.

I have no familiarity with the papers in question but TimG sounds like he knows what he is talking about in regards to them. If he is referencing other sources, he is doing it in an effective way, paraphrasing and putting them in the relevant context of the argument, rather than simply quoting long passages and spamming dozens of links like some other posters on this topic.

Edited September 13, 2011 by Bonam

[Keepitsimple]

I think that many people on the left like climate change as an issue because it gives them justifications to pass policies that they believe are good no matter what. But saying that people are motivated by their beliefs is a lot different than saying they are out to swindle people.

A good comment.

[Michael Hardner]

I agree that it's difficult to debate the science, which is why I would leave it to scientists.

I should be qualified to review the statistics here, which is why U'm still stymied by the term 'upside down' with regards to multivariate regression.

[TimG]

I don't have faith that you understand it either - you may just be quoting McIntyre here for all I know.

I am paraphrasing material I get from various sources but mostly McIntrye when it comes to proxies. I also read a few of the less rabid alarmist blogs to see what the the counter arguments are so I don't get too stuck in McIntrye's way of looking at things.

I just can't believe Mann could be that clueless. That would only be possible if there were a conspiracy.

It took me several months before I was at the point where I could believe that respected names in the climate science could be so incompetent. Unfortunately, I have studied this topic enough that I can come to no other conclusion.

That said, I do not believe that any of these climate scientists are committing deliberate fraud. The trouble history is filled with 'true believers' that have caused serious damage.

Edited September 13, 2011 by TimG

[Michael Hardner]

I am paraphrasing material I get from various sources but mostly McIntrye when it comes to proxies. I also read a few of the less rabid alarmist blogs to see what the the counter arguments are so I don't get too stuck in McIntrye's way of looking at things.

Right. So I actually took several undergraduate courses in statistical regression, so ... I'm not sure that what you're paraphrasing makes sense, and I can't ask you to explain it because you're paraphrasing it.

This is why I'm confused.

It took me several months before I was at the point where I could believe that respected names in the climate science could be so incompetent. Unfortunately, I have studied this topic enough that I can come to no other conclusion.

But you admit that you don't understand the science itself, so what is it based on ?

That said, I do not believe that any of these climate scientists are committing deliberate fraud. The trouble history is filled with 'true believers' that have caused serious damage.

I get that.