Muller reverses course in NYT op-ed

[c1ue]

Re: Muller reverses course in NYT op-ed

Mostly agree with the cut except the unkind and flippant "moron" accusation of McKibben.

We have clearly divergent beliefs on CAGW - but even disregarding that, the facts of the matter show very clearly that campaigning against CO2 emissions in the US is a complete waste of time, and that the 350 ppm 'line in the sand' is pure poppycock.

The catch is that this is not a problem at all (except to the ignorant and easily-manipulated press). Here's why:

The peer review process achieves exactly one thing; it determines whether a paper is worthy of publication. That is very different from verifying its correctness, which many non-scientists incorrectly assume its purpose to be.

When a supposedly objective scientist starts pushing out more PR than science, it is far from clear that the original label is correct.

What exactly is the difficulty to wait for peer review to complete?

Why is it necessary to periodically re-push BEST into the MSM, well before any review process has had the opportunity to function?

You may respect Muller, and I think he is an intelligent person, but the actions above are not the hallmark of science.

It should be pointed out that the wording of the statement:

In fact, said study has so far failed to pass peer review - a simple Google search will show that.

might be taken to imply that the paper was submitted for review, but rejected for publication.

The words chosen were quite clear. It is not that Muller's research won't pass review, it is that it has not yet. Any other interpretation is purely your own.

Actually, the data set itself draws no conclusion at all. It is a compilation of numerous other data sets. The group that consolidated it did draw a conclusion from it; one that Muller demonstrates is incomplete, and corrects.

That isn't strictly true. The surfacestations project was created specifically to examine the issue of siting of monitoring stations. Anthony Watts gave Richard Muller access so that he could make use of this additional information when evaluating the existing NCDC, GISS, and CRU temperature records.

Muller, for whatever reason, did not make use of the additional information offered by the surfacestations data - instead coming out with a conclusion that the existing data sets are perfectly fine.

And this might be credible were it not for the fact that the surfacestations papers are showing the complete opposite. Thus either Muller ignored the surfacestations data or he screwed up his analysis.

You decide.

Actually, this last statement is factually incorrect. Muller specifically states that his conclusion is fully supported when the higher-temperature readings associated with metropolitan areas are omitted, and only rural readings are used.

The above statement is factually correct but it omits the fact that the entire purpose of the surfacestations project was to examine both the urban heat island effect and the effects of urbanization on long term temperature records of specific sites.

Thus what Muller is saying is that if he chooses the sites which arguably haven't changed, that GISS/CRU/NCDC are correct. The problem is that surfacestations isn't about these stations, it is about the ones which have.

So what then is he saying?

[astonas]

Re: Muller reverses course in NYT op-ed

We have clearly divergent beliefs on CAGW - but even disregarding that, the facts of the matter show very clearly that campaigning against CO2 emissions in the US is a complete waste of time, and that the 350 ppm 'line in the sand' is pure poppycock.

I'll come back to this later. Just leaving it here as well to make sure that I am addressing ALL of your points, instead of cherry-picking those I feel I can make a case against, if I take them out of context. I'm on to you, C1ue. ;-)

When a supposedly objective scientist starts pushing out more PR than science, it is far from clear that the original label is correct.

Scientific objectivity is not a touchy-feely thing like objectivity is in the bloggosphere, or the newspapers. One can actually do a statistical analysis of a given data set that is supposedly random, and determine whether a bias exists in that data set, for example, that a die is loaded. (of course, attributing causality may in other cases be trickier). If the math says it has no bias (eg. a centered gaussian distribution is shown to a given statistically significant confidence level) then it doesn't have a bias, no matter how many people BELIEVE the die is loaded. And if a statistical bias does exist, then the amount of bias (be it as \delta x, or a +p(x)) can be quantified as well. So unlike in politics, or reporting, when one is a scientist, one operates in a sphere where one can on occasion actually PROVE that one is being objective. Such a proof is not always possible, of course, but sometimes it is, and it specifically is in this case.

And for this reason, your assertion, above, is not true. A person can be a publicity-hound and do good, objective, technical work, or a retiring type and do bad, biased, work, just as each can do the opposite. A scientist's results, and their objectivity, are in their numbers and equations, not just their words. If someone has the numbers and equations to back up their arguments, they can be the biggest public loudmouth or the shyest wallflower and not alter the strength of their underlying argument. It sucks, but it is true. It is possible for a jerk to be objectively right, personality notwithstanding. The key is that you have to show your numbers, and your work.

Now, not all scientists are good at being scientists, of course. And many speculate far beyond what their math will support. More on this later.

In the case of Dr. Muller, he is without a doubt, and has always been, an arrogant and grandstanding person. But his reputation as an objective scientist is still sterling in the community. This is because when he puts out work, it is always of very highest quality. And here's the important part: that's not my opinion. Nor is it just a consensus opinion in the field. It is MEASURABLY of high quality. He does the statistics, reports all his data, shows his work, links his raw data, shows error bars, gives confidence percentages, etc. In other words, he doesn't just provide data, he also proves conclusively that the data provided is both valid and complete data. So if he wants to grandstand, he can. Not because he has "earned the right to do so", or any other such nonsense. But because his work stands entirely on its own, independent of personality flaws.

His principle criticism of both the "CAGW-is-real" and the "CAGW-is-not-real" crowds is that NEITHER has achieved this level of rigor before. And he is certainly correct on that score. So when he finally came up with a fully rigorous way of doing so, of meaningfully pulling together data from all relevant places, that is immediately central to the discussion. He meets a standard of proof that both sides have thus far failed to meet.

So in his publications on this subject, when he includes data, he first PROVES that it is meaningful and proper to include it. When he excludes data, he provides statistical analysis of the methodology he uses to make that choice. His "methods" paper goes through this in great detail, and is still looking for a mathematically meaningful challenger. (ie. not just hot air from a blog)

When you show your math like that, you can't easily be accused of bias, no matter what you say publicly. Every assertion you are making has been backed up, both logically and mathematically.

It is when you don't show your math that the world has to trust you. That's when politics and polemics come in. And that's what Anthony Watts does. He relies on argument, rather than math. And that's why no one should, or in the field, does, take him seriously. He doesn't show his work. (And what he does show is embarrassingly bad.)

What exactly is the difficulty to wait for peer review to complete?

Why is it necessary to periodically re-push BEST into the MSM, well before any review process has had the opportunity to function?

Again, there is no difficulty, or necessity. As I've explained in great detail in my last post, Muller IS conducting peer review. Right now. In public. With his opponents as reviewers. This is entirely legitimate. It is only journalists, for whom the editorial process includes the additional role of fact-checking, that repeatedly misunderstand this, and pass this misunderstanding along. The scientific editorial system DOES NOT SERVE THIS FUNCTION, except perhaps as a tertiary role, or for extraordinarily bad work. The dialogue in the literature is where the actual merits of claims are ultimately evaluated. If a paper doesn't get into Science, or Nature, there is generally another, less-read journal that will eventually accept it. The difference is not publication. The difference is readership, and hence, relevance.

One group publishes a paper. Another responds. A third provides another explanation. And so on. Each time, the editor and reviewers aren't saying "this group is right". They are saying "this is interesting enough to be worth reading" to a certain community. It is the dialogue that vets the work.

In Muller's case, he can self-publish because everyone in the field already knows he's worth reading, and will read what he writes, wherever it is published. In Watts' case, he has to self-publish, because everyone knows he's not worth reading. But in both cases, anyone who wants to is free to look at what they've put out, and come to their own conclusion. What stuns me is how few people actually bother to look, before they dive in and start commenting. The difference in quality between the two is striking, and obvious.

You may respect Muller, and I think he is an intelligent person, ...

This conversation has nothing to do with my opinion of Muller. Truth be told, I think he is arrogant, and not particularly nice. But as I say, that is irrelevant to the discussion, which is about the objectivity of his work.

With respect to objectivity specifically, the complaints brought forward are simply not valid concerns, as they rest on a fundamental misunderstanding of the peer-review process. Whether I respect his objectivity or not (though I do) is still irrelevant, since the challenge criticized his approach to peer-review on a flawed understanding of what that was.

...but the actions above are not the hallmark of science.

The pursuit of a definitive conclusion that cannot be attacked by opponents as being biased (in this case, by structural weakness in the peer-review process) is EXACTLY a hallmark of science. To claim otherwise is to misunderstand the nature of science. It is about the pursuit of the objective truth, by stripping away every last opportunity for subjectivity to enter. If that subjectivity is perceived (even by a tiny subset of the community) to be in the peer-review process itself, that is precisely what must be removed from the equation for truth to be found.

That isn't strictly true. The surfacestations project was created specifically to examine the issue of siting of monitoring stations. Anthony Watts gave Richard Muller access so that he could make use of this additional information when evaluating the existing NCDC, GISS, and CRU temperature records.

Muller, for whatever reason, did not make use of the additional information offered by the surfacestations data - instead coming out with a conclusion that the existing data sets are perfectly fine.

And now we see the strong contrast between Muller and Watts (a consummate polemicist). It is in that "whatever reason." If Watts were to subject his data to the same standard Muller uses -- and he clearly never once has, just read his publications -- he would wind up proving it was, in fact, not valid data at all.

The guy's work is a disaster. There is a built-in bias in the selection of the the stations evaluated, a failure to show causality of the "flawed" stations, and on top of all that, a basic misuse of the data itself (employing momentary absolute, rather than the appropriate delta-over-time data). And that's just what I got from skimming his own self-inconsistent work just now; he doesn't even do a good job of hiding incompetence! The bottom line is that he's just too incompetent to do even an elementary statistical analysis correctly, let alone come to a conclusion that has actual meaning. I mean, have you actually looked at his "work"? It's a train wreck! It has has NONE of the properties that are required for data to be a valid input for such a project, not a valid geographical distribution, nor reproducible or reliable measurements, nor even a differentiable conclusion.

Only the lunatic fringe would take that kind of work seriously, no matter what its presentation or conclusion was. I'm sorry if you were hoodwinked by such unscrupulous folks, but that's the state of it. You were mislead, plain and simple.

The fact that NOAA has gone along with upgrading its sensors as a consequence of this guy is not a function of the strength of his data, but a demonstration of how far they are willing to bend over backwards to avoid even the appearance of bias, no matter how ridiculous the claim.

Muller was smart enough to construct standards that either exclude (or, if included piecewise, statistically underweight) bad data appropriately, based on objective standards. That is the proper procedure for handling such questionable data. No apologies needed there at all.

One challenge that this national conversation is currently experiencing is that at least one side has been informed by commentary, rather than science. This is why garbage like Watts' keeps turning up like a bad penny. Watts creates commentary, but has yet to produce one iota of actual science. He simply isn't generating statistically significant data, so no science can even be begun with it. Though clearly that hasn't stopped him from pretending to do science.

And this might be credible were it not for the fact that the surfacestations papers are showing the complete opposite. Thus either Muller ignored the surfacestations data or he screwed up his analysis.

To ignore statistically mangled results is not "screwing up" one's analysis. To include it - at all - is. Muller's process correctly rejects the garbage. Again, perfectly correct, and valid. If you have a problem with his metrics, show me where in his paper it resides. No one else has been able to yet.

You decide.

NO!!! That's the whole point of science! "You", whoever that refers to, does NOT decide. Anything. If a conclusion cannot be reached objectively, then it is not a valid conclusion. It isn't about convincing, or deciding. It is about PROVING. If no proof, then no conclusion. Here, there was no proof of either position, only assertion. So the decision is a false one.

If there is a "you" who gets to decide something, then what is happening is politics, not science. Do you see?

Watts' garbage is NOT valid. That can be proven, by measuring it against an objective, published, standard. The dataset Muller wound up including IS valid. Those aren't opinions, preferences, or decisions. Not on my part, yours, Watts', or even Muller's. That is a numerically verifiable statement, whose validity can be evaluated based on the self-consistency and applicability of objective criteria, which Muller justifies at length in his methods paper (linked above). He goes into incredible depth about how to test and process a dataset. If you wish to find a problem in that paper, by all means do so. But so far no one else has. They've just howled about "bias", as though that word means something without a standard to measure against.

They see science, and respond with politics, not understanding that the latter cannot, in the long run, alter the former at all. Reality is what it is, no matter what a politician can convince people of for a time.

The above statement is factually correct but it omits the fact that the entire purpose of the surfacestations project was to examine both the urban heat island effect and the effects of urbanization on long term temperature records of specific sites.

Thus what Muller is saying is that if he chooses the sites which arguably haven't changed, that GISS/CRU/NCDC are correct. The problem is that surfacestations isn't about these stations, it is about the ones which have.

No. Just the opposite. The surface stations project was about demonstrating that urban weather stations were misreading real temperatures, in a way which caused overestimation of global temperature increases. If this were true, then excluding those stations that were misreading would bring you back to correct values for those stations that didn't misread. Watts can't have this one both ways. Either he is lying, and there was no malfunction (making their data valid), or he is irrelevant, because there was a malfunction, and the data from them is not valid.

So what then is he saying?

(In my opinion) Muller is saying this: "Here's my (Muller's) dataset, here's my code, and here's my methods so you don't even have to read and understand the raw code. Find a real problem in one or more of these, and we can talk again. Unless someone can prove such a problem exists, the most rigorous and complete study conducted to date shows that CAGW is real, and the dominant factor associated with GW today. The work stands on its own merits. Now come try to tear it down, and we'll see if I'm right."

Everything else is ego-stroking, on the part of one person or another.

Oh, and I said I'd come back to this one:

We have clearly divergent beliefs on CAGW - but even disregarding that, the facts of the matter show very clearly that campaigning against CO2 emissions in the US is a complete waste of time, and that the 350 ppm 'line in the sand' is pure poppycock.

I know that this was addressed to santafe2, and not me. But the "facts" show nothing of the kind. If anything, they show just how far TV meteorologists like Watts have to stretch in order to make a case. It saddens me, and even more so when intelligent people believe their incredibly thinly-supported misrepresentations.

p.s.

I think I'm done with this topic for now. I'm sure you will want to respond, but I rather suspect that it will be to argue points, rather than develop a deeper understanding of the issue. Please feel free to take the last word, if that is what you seek.

[c1ue]

Re: Muller reverses course in NYT op-ed

The key is that you have to show your numbers, and your work.

No disagreement here. The problem is that the consensus doesn't like anyone outside of the circle to see the numbers, as has been very well documented by McIntyre and others.

His principle criticism of both the "CAGW-is-real" and the "CAGW-is-not-real" crowds is that NEITHER has achieved this level of rigor before. And he is certainly correct on that score. So when he finally came up with a fully rigorous way of doing so, of meaningfully pulling together data from all relevant places, that is immediately central to the discussion. He meets a standard of proof that both sides have thus far failed to meet.

I agree that objective measurement is the hallmark of a scientist, I would note that Dr. Muller's sturm und drang over BEST has not been accompanied thus far by anything of particular note.

There is certainly value is validating other's work, but I fail to see what objectively that's been done so far with BEST that is so much to Muller's credit.

What he's said is that he agrees with the consensus, which he was already inclined to anyway.

As for rigor - the reality is that neither side can show rigor.

Climate is a decade to millenial phenomenon, and for anyone including Muller to say that a definitive understanding of climate can be obtained in a 3 decade old field is, frankly, ludicrous.

It is when you don't show your math that the world has to trust you. That's when politics and polemics come in. And that's what Anthony Watts does. He relies on argument, rather than math. And that's why no one should, or in the field, does, take him seriously. He doesn't show his work. (And what he does show is embarrassingly bad.)

I'd say that given that Watts isn't trying to be a climate scientist, and furthermore has been completely open to allowing anyone to access the surfacestations data, that your criticism is pretty far off the mark.

You can decry this style all you want, but ultimately he doesn't seek to muzzle anyone and in fact actively encourages debate over climate.

In Muller's case, he can self-publish because everyone in the field already knows he's worth reading, and will read what he writes, wherever it is published. In Watts' case, he has to self-publish, because everyone knows he's not worth reading. But in both cases, anyone who wants to is free to look at what they've put out, and come to their own conclusion. What stuns me is how few people actually bother to look, before they dive in and start commenting. The difference in quality between the two is striking, and obvious.

...

This conversation has nothing to do with my opinion of Muller. Truth be told, I think he is arrogant, and not particularly nice. But as I say, that is irrelevant to the discussion, which is about the objectivity of his work.

With respect to objectivity specifically, the complaints brought forward are simply not valid concerns, as they rest on a fundamental misunderstanding of the peer-review process. Whether I respect his objectivity or not (though I do) is still irrelevant, since the challenge criticized his approach to peer-review on a flawed understanding of what that was.

The problem with what you're saying is that if Muller really believes in public peer review, then what is the point of private peer review?

You're essentially betting on the man - and that's your prerogative.

Unfortunately your sterling view of Muller is not shared by many other in the climate community - not merely the flamers on both sides but also those who are much more analytic and nuanced:

It’s particularly notable that one collaborator on the first batch of papers, Judith Curry of the Georgia Institute of Technology, declined to be included as an author on the new one. I learned this when I sent her this question by e-mail:

Do you share Rich’s extremely high confidence on attribution of recent warming to humans…?

Here’s Curry’s reply:

I was invited to be a coauthor on the new paper. I declined. I gave them my review of the paper, which was highly critical. I don’t think this new paper adds anything to our understanding of attribution of the warming….

I really like the data set itself. It is when they do science with it that they get into trouble.

Curry also sent this note, which she is distributing to other journalists:

The BEST team has produced the best land surface temperature data set that we currently have. It is best in the sense of including the most data and extending further back in time. The data quality control and processing use objective, statistically robust techniques. That said, the scientific analyses that the BEST team has done with the new data set are controversial, including the impact of station quality on interpreting temperature trends and the urban heat island effect.

Their latest paper on the 250-year record concludes that the best explanation for the observed warming is greenhouse gas emissions. Their analysis is way oversimplistic and not at all convincing in my opinion.

There is broad agreement that greenhouse gas emissions have contributed to the warming in the latter half of the 20th century; the big question is how much of this warming can we attribute to greenhouse gas emissions. I don’t think this question can be answered by the simple curve fitting used in this paper, and I don’t see that their paper adds anything to our understanding of the causes of the recent warming. That said, I think there are two interesting results in this paper, regarding their analysis of 19th century volcanoes and the impact on climate, and also the changes to the diurnal temperature range.

From my perspective as a longtime, but lay, analyst of climate science, my sense is she has it right. The data-sifting methods of the Berkeley project, largely developed by a brilliant data analyst, Robert Rohde (there’s more on him here), have clearly added value to longstanding efforts to clarify temperature trends across a variegated planet. But the conclusions Muller describes now do seem overly simplistic (as Curry, Connolley and others say).

So, a kudos to Rohde, the opposite for Muller.

The pursuit of a definitive conclusion that cannot be attacked by opponents as being biased (in this case, by structural weakness in the peer-review process) is EXACTLY a hallmark of science. To claim otherwise is to misunderstand the nature of science. It is about the pursuit of the objective truth, by stripping away every last opportunity for subjectivity to enter. If that subjectivity is perceived (even by a tiny subset of the community) to be in the peer-review process itself, that is precisely what must be removed from the equation for truth to be found.

Unfortunately, it seems quite clear that Muller's conclusions are far from unassailable from any direction. See above.

And now we see the strong contrast between Muller and Watts (a consummate polemicist). It is in that "whatever reason." If Watts were to subject his data to the same standard Muller uses -- and he clearly never once has, just read his publications -- he would wind up proving it was, in fact, not valid data at all.

The guy's work is a disaster. There is a built-in bias in the selection of the the stations evaluated, a failure to show causality of the "flawed" stations, and on top of all that, a basic misuse of the data itself (employing momentary absolute, rather than the appropriate delta-over-time data). And that's just what I got from skimming his own self-inconsistent work just now; he doesn't even do a good job of hiding incompetence! The bottom line is that he's just too incompetent to do even an elementary statistical analysis correctly, let alone come to a conclusion that has actual meaning. I mean, have you actually looked at his "work"? It's a train wreck! It has has NONE of the properties that are required for data to be a valid input for such a project, not a valid geographical distribution, nor reproducible or reliable measurements, nor even a differentiable conclusion.

That's funny, because surfacestations isn't Watts work. It is a crowdsourcing project for which the results are available to anyone who asks.

Thus your opinion is clearly based on a series of flawed assumptions.

Furthermore the papers coming out of the surfacestations project are undergoing the exact same 'peer review' process, so the truth will out whatever your or my opinion may be.

From my point of view, I haven't been commenting on surfacestations exactly because it was far too early to do so. However, that is no longer true, and you can see in Climate Change where things are at now.

Only the lunatic fringe would take that kind of work seriously, no matter what its presentation or conclusion was. I'm sorry if you were hoodwinked by such unscrupulous folks, but that's the state of it. You were mislead, plain and simple.

I'm glad you are saying so, because it shows clearly what your biases are.

Roger Pielke Senior - a prominent warmist (i.e. believes in AGW but not in the consensus), a former CCSP member, a scientist who does not grandstand nor promote any policy position whatsoever, thinks differently.

To ignore statistically mangled results is not "screwing up" one's analysis. To include it - at all - is. Muller's process correctly rejects the garbage. Again, perfectly correct, and valid. If you have a problem with his metrics, show me where in his paper it resides. No one else has been able to yet.

The problem is, Muller didn't "ignore" it. He simply failed to do any meaningful analysis of it despite loud public claims of taking this new data into account.

NO!!! That's the whole point of science! "You", whoever that refers to, does NOT decide. Anything. If a conclusion cannot be reached objectively, then it is not a valid conclusion. It isn't about convincing, or deciding. It is about PROVING. If no proof, then no conclusion. Here, there was no proof of either position, only assertion. So the decision is a false one.

Wrong again. Because as I noted above - the time scale of the phenomena in question as well as the provenance and consistency of the records to date - are such that any and every conclusion can at best be termed hypothetical.

The bar to change from hypothetical to theoretical is simple: empirical demonstration of predictive skill.

So far, nothing. If anything, CAGW predictions are more consistently wrong than right - which is skill of a sort but not the one you normally want.

No. Just the opposite. The surface stations project was about demonstrating that urban weather stations were misreading real temperatures, in a way which caused overestimation of global temperature increases. If this were true, then excluding those stations that were misreading would bring you back to correct values for those stations that didn't misread. Watts can't have this one both ways. Either he is lying, and there was no malfunction (making their data valid), or he is irrelevant, because there was a malfunction, and the data from them is not valid.

For someone who is supposedly a scientist, you don't seem to have understood what the surfacestations project is actually about. It seems your understanding is informed far more by the AGW propagandists than by fact.

Surfacestations is not about erroneous readings - i.e. lack of skill or equipment induced error.

Surfacestations is about how the siting of temperature record taking locations is poor (i.e. fails to meet published guidelines) and has changed over time - such that many locations today are mislabeled (rural vs. semi-rural vs. urban), are sampling temperatures in non-representative settings (i.e under an air conditioner), etc etc

The hypothesis is that the influence on temperatures by these errors is unidirectionally up - not random, and therefore if there are a large number of these stations, that there would be a unidirectionally upward bias effect to the overall temperature record.

A secondary question is just how valid the various NCDC/GISS/CRU/GHCN programs validate and equalize the temperature records at any given location over time - because many of these stations go offline, online, get moved, or have some other major change. There are a number of inconsistencies in this area as well, but it is not a primary surfacestations focus.

(In my opinion) Muller is saying this: "Here's my (Muller's) dataset, here's my code, and here's my methods so you don't even have to read and understand the raw code.

Except it isn't his data set, it is Rohde's. And his analysis has been slammed from one side of the spectrum to the other as noted above.

I know that this was addressed to santafe2, and not me. But the "facts" show nothing of the kind. If anything, they show just how far TV meteorologists like Watts have to stretch in order to make a case. It saddens me, and even more so when intelligent people believe their incredibly thinly-supported misrepresentations.

The facts show that the US is no longer the dominant emitter of CO2, nor is the developed world. Are your facts different?

[santafe2]

Re: Muller reverses course in NYT op-ed

The facts show that the US is no longer the dominant emitter of CO2, nor is the developed world. Are your facts different?

Not many could be in disagreement on this point. We've offshored our jobs and our impact on the earth to the least sustainable country. That said, all humans are accelerating our pace toward an environmental tipping point. I'm not sure how making an intellectual slam dunk matters in the larger view.

[don]

Re: Muller reverses course in NYT op-ed

Not many could be in disagreement on this point. We've offshored our jobs and our impact on the earth to the least sustainable country. That said, all humans are accelerating our pace toward an environmental tipping point. I'm not sure how making an intellectual slam dunk matters in the larger view.

I'm a native of upstate New York. My Dad, years after I left that rustbelt, told me it was amazing how the lakes and rivers were returning to health now that the factories were gone. 'Nobody' had work but the air was fresh.

The sheeple, sucker punched again . . . .