GLOBAL WARMING BOMBSHELL: Hockeystick Broken

A prime piece of evidence linking human activity to climate change turns out to be an artifact of poor mathematics.

Progress in science is sometimes made by great discoveries. But science also advances when we learn that something we believed to be true isn't. When solving a jigsaw puzzle, the solution can sometimes be stymied by the fact that a wrong piece has been wedged in a key place.

In the scientific and political debate over global warming, the latest wrong piece may be the "hockey stick," the famous plot (prominently displayed by the IPCC report, 2001), published by University of Massachusetts geoscientist Michael Mann and colleagues. This plot purports to show that we are now experiencing the warmest climate in a millennium, and that the earth, after remaining cool for centuries during the medieval era, suddenly began to heat up about 100 years ago--just at the time that the burning of coal and oil led to an increase in atmospheric levels of carbon dioxide.

I talked about this at length in my December 2003 column. Unfortunately, discussion of this plot has been so polluted by political and activist frenzy that it is hard to dig into it to reach the science. My earlier column was largely a plea to let science proceed unmolested. Unfortunately, the very importance of the issue has made careful science difficult to pursue.

But now a shock: independent Canadian scientists Stephen McIntyre and Ross McKitrick have uncovered a fundamental mathematical flaw in the computer program that was used to produce the hockey stick. In his original publications of the stick, Mann purported to use a standard method known as principal component analysis, or PCA, to find the dominant features in a set of more than 70 different climate records.

But it wasn't so. McIntyre and McKitrick obtained part of the program that Mann used, and they found serious problems. Not only does the program not do conventional PCA, but also it handles data normalization in a way that can only be described as mistaken.

Now comes the real shocker. This improper normalization procedure tends to emphasize any data that do have the hockey stick shape, and to suppress all data that do not. To demonstrate this effect, McIntyre and McKitrick created some meaningless test data that had, on average, no trends. This method of generating random data is called "Monte Carlo" analysis, after the famous casino, and it is widely used in statistical analysis to test procedures. When McIntyre and McKitrick fed these random data into the Mann procedure, out popped a hockey stick shape!

That discovery hit me like a bombshell, and I suspect it is having the same effect on many others. Suddenly the hockey stick, the poster-child of the global warming community, turns out to be an artifact of poor mathematics. How could it happen? What is going on? Let me digress into a short technical discussion of how this incredible error took place.

In PCA and similar techniques, each of the (in this case, typically 70) different data sets have their averages subtracted (so they have a mean of zero), and then are multiplied by a number to make their average around that mean to be equal to one; in technical jargon, we say that each data set is normalized to zero mean and unit variance. In standard PCA, each data set is normalized over its complete data period; for the global climate data that Mann used to create his hockey stick graph, this was the interval 1400-1980. But the computer program Mann used did not do that. Instead, it forced each data set to have zero mean for the time period 1902-1980, and to match the historical records for this interval. This is the time when the historical temperature is well known, so this procedure does guarantee the most accurate temperature scale. But it completely screws up PCA. PCA is mostly concerned with the data sets that have high variance, and the Mann normalization procedure tends to give very high variance to any data set with a hockey stick shape. (Such data sets have zero mean only over the 1902-1980 period, not over the longer 1400-1980 period.)

The net result: the "principal component" will have a hockey stick shape even if most of the data do not.

McIntyre and McKitrick sent their detailed analysis to Nature magazine for publication, and it was extensively refereed. But their paper was finally rejected. In frustration, McIntyre and McKitrick put the entire record of their submission and the referee reports on a Web page for all to see. If you look, you'll see that McIntyre and McKitrick have found numerous other problems with the Mann analysis. I emphasize the bug in their PCA program simply because it is so blatant and so easy to understand. Apparently, Mann and his colleagues never tested their program with the standard Monte Carlo approach, or they would have discovered the error themselves. Other and different criticisms of the hockey stick are emerging (see, for example, the paper by Hans von Storch and colleagues in the September 30 issue of Science).

Some people may complain that McIntyre and McKitrick did not publish their results in a refereed journal. That is true--but not for lack of trying. Moreover, the paper was refereed--and even better, the referee reports are there for us to read. McIntyre and McKitrick's only failure was in not convincing Nature that the paper was important enough to publish.

It certainly does not negate the threat of a long-term global temperature increase. In fact, McIntyre and McKitrick are careful to point out that it is hard to draw conclusions from these data, even with their corrections. Did medieval global warming take place? Last month the consensus was that it did not; now the correct answer is that nobody really knows. Uncovering errors in the Mann analysis doesn't settle the debate; it just reopens it. We now know less about the history of climate, and its natural fluctuations over century-scale time frames, than we thought we knew.

If you are concerned about global warming (as I am) and think that human-created carbon dioxide may contribute (as I do), then you still should agree that we are much better off having broken the hockey stick. Misinformation can do real harm, because it distorts predictions. Suppose, for example, that future measurements in the years 2005-2015 show a clear and distinct global cooling trend. (It could happen.) If we mistakenly took the hockey stick seriously--that is, if we believed that natural fluctuations in climate are small--then we might conclude (mistakenly) that the cooling could not be a natural occurrence. And that might lead in turn to the mistaken conclusion that global warming predictions are a lot of hooey. If, on the other hand, we reject the hockey stick, and recognize that natural fluctuations can be large, then we will not be misled by a few years of random cooling.

A phony hockey stick is more dangerous than a broken one--if we know it is broken. It is our responsibility as scientists to look at the data in an unbiased way, and draw whatever conclusions follow. When we discover a mistake, we admit it, learn from it, and perhaps discover once again the value of caution.

Richard A. Muller, a 1982 MacArthur Fellow, is a physics professor at the University of California, Berkeley, where he teaches a course called "Physics for Future Presidents." Since 1972, he has been a Jason consultant on U.S. national security.

The link you posted is not a peer-reviewed paper, just an opinion piece. Also, since it was posted 2 years before the paper in question, I have doubts about how relevant it is. It was posted 2 years earlier. However I do note that it still quotes from Keigwin (see my comments re his work on the Sargasso Sea above).

The quote I gave was directly from their methodology. Let me repeat it again.

"Table 1 and Figs. 1 to 3 summarize the answers to the
questions posed here about local climatic anomalies.
For Questions (1) and (2), we answered ‘Yes’ if the
proxy record showed a period longer than 50 yr of
cooling, wetness or dryness during the Little Ice Age,
and similarly for a period of 50 yr or longer for warming,
wetness or dryness during the Medieval Warm
Period."

That is a direct quote from their paper. So are you saying that it is sound methodology to count a period of both wet and dry as an indication of both warming and cooling?

"I didn't see you address the issue of Mann's deducing temperature from ice cores, tree rings, etc for the 1000 to 1900 part of the graph and using measurements for 1900 onwards"

Give me a break!! I am only one person and each of my replies is composed by me (except as noted). This is more time consuming than just cutting and pasting from a website. For example, post number 83 is a cut and paste directly from CO2 Science. I had to actually try to dig out the paper and when that failed I had to try and see what other work the authors had done so I could get an idea of what they said as opposed to how CO2 Science put the spin on it. Going back to source papers is time consuming but I like to do my own thinking.

In regards to Mann, the way he matched up the records was (and this is a direct quote from Mann's paper):

"We first decompose the twentieth-century instrumental data into its dominant patterns of variability, and subsequently calibrate the individual climate proxy indicators against the time histories of these distinct patterns during their mutual interval of overlap."

The method he used to decompose the instrumental data is the Principle Component Analysis (that is the current area of discussion between McKitrick and Mann). Again, from the paper:

"We isolate the dominant patterns of the instrumental surface temperature data through principal component analysis (PCA). PCA provides a natural smoothing of the temperature field in terms of a small number of dominant patterns of variability or ‘empirical eigenvectors’."

I do not agree with your last comment. TO me (from what I understand) it looks like fairly solid science to me. However it does appear (to me anyway) to be fairly complex. You might have the math necessary so that this appears simple, I do not. However I am always willing to learn.

"Also, I am very happy to accept that the MWP and LIA did exist. I don't know of anyone who says otherwise."

Okay, point of agreement... excuse my hasty previous reply.

If you accept LIA, though, it's simply a rabbit trail to dispute the underlying cause of a particular temperature record locally. Daly's argument is not about why they changed, but about the existence of change itself in the previous millenium. If you accept it (LIA), you accept his point.

It have been noted that CO2 fertilization affects growth rates, in turn biasing the tree ring records to disattenuate prior temperature variability in the record. Since Mann over-emphasized in his PCA the bristlecone pine samples, this effect may have impacted his results. You can nitpick sample after sample this way.

"I believe I have discussed most (all?) of them and have shown that they do not support the conclusions as presented."

Your view.

"What I object to is poor scientific methodology and biased reporting being introduced into the debate. "

Then you should be red in the face over the IPCC reports and their phony exagerrated claims, over the data errors known to be in MBH98 that were only corrected 6 years later (and perhaps not fully acknowledged yet with Mann instead attacking M+M rather than fixing his work or making it more reproducible by others), and the 'reporting' of the media that hypes the threat and the politized science in the area.

However it does appear (to me anyway) to be fairly complex. You might have the math necessary so that this appears simple, I do not. However I am always willing to learn.

Your earlier appeal to authority has now turned into an appeal of ignorance, an interesting change of rhetoric.

"We isolate the dominant patterns of the instrumental surface temperature data through principal component analysis (PCA). PCA provides a natural smoothing of the temperature field in terms of a small number of dominant patterns of variability or `empirical eigenvectors'."

Let's see if we can figure this out together. An vector is a measurement of each dimension of orthogonal values, such values would be measurements such as tree ring width, temperature, etc. Vectors denote a point in the space made up of those axes, that point relates the various measurements. For example a wider tree ring combined with a higher temperature means in practical terms that the tree grew more in a warmer year. As stated by Mann and his critics, the only way such relationships can be determined is by using temperature measurements made in the 1900-2000 century. Then Mann applied those relationships to the previous known measurements (tree rings, ice cores and one other that I forgot).

The potential problems with this method are numerous. Foremost, the data can be cherry picked. Apparantly Mann used just nine locations out of all the available data, just 5 for North America. Surely much more data could have been used. Second, other factors could skew the relationship between the measurements. In the case of tree rings there is certainly a man-made component of increased CO2 that would cause increased tree growth for a given temperature. Third, other factors can easily be ignored, How would sun intensity affect plant growth, how would that be measured in history and how would that relate to temperature (as we must certainly agree it does)?

I do not have the papers in front of me so I can only provide some general comments. If you have links to them I will be happy to review and respond in detail.

Argentinian Glaciers. How a glacier grows is related to temperature and precipitation. So how did Soon correct for precipitation effects? Without knowing this it can not be used as a temperature proxy.

Chinese Cultivation. That paper actually is more concerned with how solar output changes climate (specifically as measured by oxygen isotopes and carbon isotopes). It does not appear to give actual temperatures but ties the solar record to historic climate records previously published.

Finally, regarding stalagmite composition, these are sensitive to precipitation and geological and visitation changes and thus they must be used with caution. However without the paper I can't say if the author corrected for these.

Now, I have answered your question, could you please address mine. Do you feel that Soon's methodology as presented in Soon's paper posted by WOSG is scientifically sound?

There are a number of hypotheses in the paper. The most explicit one is about measured climate anomalies correlating with LIA or MWP. As you point out, the anomaly correlation can be cooling OR dryness OR wetness in the LIA and warming OR dryness OR wetness in the MWP. But the paper also proposes hypotheses for a shift from wet to dry or from dry to wet correlating with cooling or warming. The text had some explanations for these occurances but the tables only had the occurances which could admittedly point either way. But the conclusions are separate, that the 20th century is not the warmest, nor the most climatically extreme.

In my opinion Both conclusions are supported by the data. Clearly the climatic extreme question is resolved beyond a doubt, there are far fewer in the 20th century than in the previous 11 centuries. The warming and cooling is supported less fully, but still quite generally. The difficulty of correlating the proxies to actual warming and cooling is discussed as are many possible local effects that correlate wetness or drying to cooling or warming. In many cases warmer meant wetter and cooler meant drier as explained by plausible local phenomena. There were exceptions that were also fairly well explained.

I would say the two conclusions are hard to separate because of the relationship of both types of measurements (temp and precip) to the proxies. But the implication that dryness and wetness correlates with temperature changes is well supported. And so then is the conclusion that the 20th century is not extreme in either warming or cooling. As for warming in MWP and cooling in LIA, they seem to be well supported although not as clearly as their being extremes of dryness or wetness. The proxies clearly indicate that, but not as clearly the temperature. The temperature conclusions are obvious for Europe and N. America, but less for the rest of the world. But there's little evidence for the rest of the world that the LIA was warm or that the MWP was cool anywhere.

LOL, sorry, no rhetoric and I have no qualms about admitting that I don’t know (that is the start of all wisdom).

Thanks for the comment on vectors, however it was not really all that useful. So if I can describe the process as I understand it. Mann took the instrumentation data and decomposed it into it main components of variability (the Eigen vectors). He then used the proxy series and analyzed these in terms of the components of variability. This gave him the influence of each component.

You then said “Apparently Mann used just nine locations out of all the available data, just 5 for North America.”

No, that is not correct. First, it depends on which paper you are talking about. In his 98 paper that looked at the global temperature he used over 100. In his 99 paper he looked at just the Northern Hemisphere and he used 12 since that is all that were available. However the first 3 were actually the 3 principle components of the 28 tree series for North America so if you consider these as well there are almost 40.

In regards to your list of potential problems with tree rings, these are taken into account. I have explained why there were just 12. In regards to CO2 enhancement, if you have read his paper you will see that he has dealt with CO2 enhancement. He also looks at solar output and checks his data series for dependence. He then removes these “non-climatic” factors from his series.

I don't think your explanation is all that different from mine. What good is a "component of variability" if it is not or can't be measured? I should have mentioned the fact that averaging the vectors gives the influence of each component.

Your comment about CO2 and sun intensity seem plausible (I haven't read Mann's explanation) since they would probably not affect the proxy measurements as much as the climate itself and local affects like cloudiness (to throw another one at you). Mann's 98 paper starts with the LIA so it is not really very useful. Don't you find it interesting that to extend the previous 6 centuries to 11 he had to discard a lot of data? It was certainly available to other researchers as pointed out in Soon's paper.

The biggest problem is your hypothesis on this thread that the climate changes were local to the North Atlantic. That local variation is not addressed by Mann with corresponding warming in other parts of the planet during the same time.

Humm, so what makes a drought in a region of the world indicate a warming during the MWP while an identical drought in the identical place indicates a cooling in the LIA? The paper does comment on the shift from wet to dry but does not offer any reason at to why it is true and in fact uses what ever data is handy to support the conclusions regardless of whether it is wet or dry.

For example to support his position he uses the following. In central Argentina, he describes the climate as more humid with increased lake sizes during the MWP. For the North Atlantic he describes the effects of the LIA as being unusually cold and wet. And in Europe, he describes flooding during the transition period between the MWP and the LIA. So increase in wetness can mean the Medieval Warm Period, the Little Ice Age and even the time in between!!! If you want to look at dry conditions, in China extremely dry conditions prevailed during the 16th and 17th centuries. In South America he states that the Patagonia region was abnormally dry during the MWP.

In most cases he presents them with no indication of why they support this particular conclusion. However when he does present a reason, it is pulled out of a hat because it would seem to match his data without any thought given to other possibilities or reasons. For example in discussing the North American west coast he notes that California, the NW Great Basin, and the northern Rocky Mountains/Great Plains experienced drought, then a period of wet, then back to a period of dry. You might think that this would cause problems for his wet = warm hypothesis, but no, the changes were caused by “the contraction and subsequent expansion of the circumpolar vortex.” If there is any justification for this besides the fact that it fits his theory, I can’t find it!

The very best that this paper can do is to show that there were weather anomalies at certain times. He does present some temperature information, but even this is botched. He used the same deMenocal and Keigwin papers as discussed before. Both of these are presented as evidence for the LIA and MWP when in fact what they show was a change in ocean circulation. Now if you want to argue that the change in ocean circulation was caused by climatic conditions consistent with the LIA and MWP, then that is a different and better argument, but he does not argue this and they are taken as temperature proxies.

So, contrary to you I feel that the paper is not well documented and not well supported. On the other hand it does have a heck of a reference section which I know impresses some.

Good night.
Yelling.

For example to support his position he uses the following. In central Argentina, he describes the climate as more humid with increased lake sizes during the MWP. For the North Atlantic he describes the effects of the LIA as being unusually cold and wet.

Reasonable questions, although these aren't "identical" places as you state. Your skepticism matches mine when I wonder where Mann documents the counterbalancing warmth that would be needed to show the whole world didn't cool down during the LIA. And as you dig it out, consider that it could appear to be just as "pulled out of a hat" as Soon's polar vortex changes (which you seem to have no interest in researching).

He does present some temperature information, but even this is botched. He used the same deMenocal and Keigwin papers as discussed before.

And all his other temperature information? All botched? You admit his hypothesis of precip extremes (dry/wet) and local temperature anomalies, but the same evidence fits nicely into global temperature anomalies if you just admit the reality of local effects. Yes, the polar vortices do mean cold=wet or warm=wet depending on location. Global temp changes will always vary by location.

Global temperature variations caused by the sun are the best explanation for the large numbers of climate anomalies clustered in the LIA and MWP, and it is clear that there where nowhere near as many anomalies in the past century.

Re: identical places, granted, I was just trying to illustrate the point.

Re: Mann, I have never understood Mann to say that.

In his 98 paper he says " ... the mid-seventeenth to early eighteenth century which corresponds to an especially cold period."

In his 99 paper he says "Our reconstruction thus supports the notion of relatively warm hemispheric conditions earlier in the millennium, while cooling following the 14th century could be viewed as the initial onset of the Little Ice Age"

Also if you look at Mann's reconstruction you see that he shows a cooling of about 0.8C where as the papers that Soon cites seem to call for a cooling of between 1 and 2C. This is well within the error bars on Mann's work (I couldn't find any error bars associated with Soon's work). So, no I don't think Mann just pulled it out of a hat.

Re: Polar Vortex, hey, I wasn't the one that wrote it! If there is any researching/explaining I would expect it to come from Dr. Soon. Also, I don't see how he can make the link from Global temperature anomaly to climate anomaly.

You also say "You admit his hypothesis of precip extremes (dry/wet) and local temperature anomalies,". I said that this is what he says. I don't accept it as a general rule. It may mean warming, it may mean cooling or it may mean no temperature change.

Re: His other temperature information, I don't know. Some I think is correct (i.e. Mann but I note that he even botched the reference to Mann), some I will try to read up more on (i.e. there appears to be an interesting one on China I haven't read). I think that if he had left his research at dealing with these he would have had a much stronger paper.

Re: anomalies this century, I don't agree with you on this. I think the anomaly's are there but are reported different. Anyway, I would need to see more than Soon's list to confirm this in my mind.

Finally, you call me skeptical and that is true as I suspect it is of you. I also think that skeptic is a complement. A skeptic who thinks about things is an interesting person who can be trusted to make up their own mind and I hate sheep (give me a skeptic who's wrong over a sheep who's right any-day).

Also, I don't see how he can make the link from Global temperature anomaly to climate anomaly.

They are essentially the same thing. A global temperature anomaly manifests as a much larger-than-normal number of local climate anomalies. Global cooling manifests as a large number of local coolings and locally above or below normal precip. Likewise warming to warming and abnormal precip. The other scientists' explanations that he reviews are from local weather pattern changes (e.g. the polar vortex). The hypothesis I allude to in 138 is that global temperature anomalies are more likely to create climate anomalies than when global temperatures are normal. I think that is well supported both empirically and in theory.

I think that if he had left his research at dealing with these he would have had a much stronger paper.

Perhaps so, the precip changes are clearly pretty strong for both periods but the explanations are rather piecemeal and some are non-existant. Part of that is the complexity of local climate changes and partly he is just ignoring the issue.

Re: anomalies this century, I don't agree with you on this. I think the anomaly's are there but are reported different.

That is the crux of the issue. Do the proxy measurements from the 1900's indicate greater warmth than the MWP? Soon points out over and over that the answer is no. There is no conceivable way that so many proxy measurements could all be distorted by precipitation. Some are more likely (glaciers) and some less likely (tree line). Some have better explanations (ocean temperatures from changes in ocean currents). But the overwhelming evidence from Soon's paper is for warming in the MWP well beyond any in the 20th century (and most of that was early 20th in any case).

Global cooling manifests as a large number of local coolings and locally above or below normal precip. Likewise warming to warming and abnormal precip.

But you still need to link the change in precipitation to either warming or cooling at that time in that area. If you can't do this, the information is worthless.

The other scientists' explanations that he reviews are from local weather pattern changes (e.g. the polar vortex).

But again, there is nothing to link a change in the polar vortex to temperature change. All the polar vortex will do is to change the circulation air which WILL change precipitation.

The hypothesis I allude to in 138 is that global temperature anomalies are more likely to create climate anomalies than when global temperatures are normal. I think that is well supported both empirically and in theory.

I am not as sure of the connection as you are, however it is not really relevant here. The key that I keep going over and over is that you can not just point to anomalies and say that they are indications of warming or cooling. There must be a way to tie the anomaly's to temperature change and then to either warming or cooling.

But the overwhelming evidence from Soon's paper is for warming in the MWP well beyond any in the 20th century (and most of that was early 20th in any case).

Absolutely not! As discussed above, we can't say what the changes in precipitation show in regards to temperature. Also, few if any of the studies are designed to look for 20th century anomalies. These are all (as far as I can see) papers trying to study past events and they use appropriate methods for that purpose. For example, ice core samples depend on the depth of the firn ice which can take up to several hundred years to develop and radio-carbon dating is only good after about 50 years has passed and will not generally give you a reading within much less than 100 years. Add to this the error of measurement such as (quoting from Bond, another author Soon references) "We sampled both cores at intervals of 0.5 to 1 cm (equivalent to a resolution of 50 to 100 years)" So if the resolution is 50 to 100 years, can we say anything about the last 1/2 of the 20th century?

Finally, I will add to the list of errors that Soon has made in this paper by saying that he or his researchers has a reading comprehension problem (using WOSG's terminology). I was re-reading deMenocal and I will note that Soon has listed him as answering NO to the question of "Is there an objective discernible climatic anomaly within the 20th century?" (as given in Table 1 - deMenocal 2000). In his paper deMonocal says " ... the LIA, ended in the late 19th century, and some of the warming since that time may be related to the present warming phase of this millennial-scale climate oscillation (Fig. 4), although the warming in recent decades is unprecedented relative to the past millennium." Of course he is talking about warming due to changing currents but that didn't stop Soon before.!

Absolutely not! As discussed above, we can't say what the changes in precipitation show in regards to temperature. Also, few if any of the studies are designed to look for 20th century anomalies. These are all (as far as I can see) papers trying to study past events and they use appropriate methods for that purpose.

Not absolutely anything! There are plenty of proxies that are just temperature. The only debate is the global extent of warming in the MWP. I researched the examples you stabbed at above and found it wasn't just glaciers in Argentina (with possible precip influence) but climate compared to today (Cioccale 99). The Chinese example wasn't just cultivation but oxygen isotopes that are determined by temp, not precip. The stalagmites are affected by both temp and precip but the precip can be controlled (www.gsf.fi/esf_holivar/holmgren.pdf)

Admittedly, the data for global warming in MWP is not as strong as the North Atlantic, but it exists in the South Atlantic and South America. The others are wet and dry, or may have precip influence (e.g. Aussie tree rings). The Peruvian glacier shows some hockey stick characteristics, but also that today's temperatures are a bit cooler than the MWP. That one is certainly a temperature proxy.

In short, the evidence for MWP warmth is substantial but doesn't exceed today's temperatures by much except for N America and Europe. It was certainly global in extent and does not support the contention that the 20th century was warmer. At best it was similar in the early part of the century.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.