I fully agree – but this also means that there is no way of knowing whether dendro data have re constructive value as a proxy either. This is the reason that ‘the decline’ in proxy temperatures had to be hidden – not just to make a pretty graph. There must also be doubt over the CO2 ‘perfect match’ between ice core data taken at the pole where CO2 atmospheric concentrations are always lowest and the current measurements supposedly a world average. One wonders whether any ‘decline’ has been hidden there.

“The distrust that many of us who read Anthony’s blog have is that the primary evidence (viz the surface station data) is in the hands of people who believe that the message is more important than the sciencle and who have manifestly manipulated data to reinforce it.”

Again this is an extremely important point. Surely it is time that metrics and their validation were separated from the research into what these metrics may mean. A UN organization cannot be expected to respect academic neutrality but scientists should.

“Ed asserts that we know that the actual temperature readings were higher than projected. That’s exactly the point, Ed. We don’t.”

As it is now – no scientist now knows who or what to trust; decades of cited peer reviewed research and learning may be based on unsound premises.

Substituting the figures from what actually happened was the right thing to do. It showed what actually happened, and cutting off the dendro proxy data prevented the presentation of a much more erroneous projection to the future.

What seems impossible to get across in this discussion is that the only real science produced in these papers is a categorical demonstration that dendro data have no predictive value as a proxy for near surface temperature. That’s an interesting and valuable finding. Why not leave it at that?

What was actually published was scriptoral. It is a conflation of whatever straws were to hand to illustrate a thesis. This is not science and would get short shrift in any other hard science discipline. The distrust that many of us who read Anthony’s blog have is that the primary evidence (viz the surface station data) is in the hands of people who believe that the message is more important than the science and who have manifestly manipulated data to reinforce it.

Ed asserts that we know that the actual temperature readings were higher than projected. That’s exactly the point, Ed. We don’t.

I’m now convinced that your view of reality is hopelessly distorted. Say whatever you like, as it has no impact. But the fact remains that a portion of the tree ring dataset that showed a decline in temperature post 1961 was combined with data that showed the opposite result. If a stock report did something like this for a major corporation stock trend, people would go to jail for defrauding investors.

Make your case for claiming a decline after 1960, then. Remember, this chart was published in the 1990s. We knew what happened in the 1960s, 1970s, and 1980s. To do as you say, and show a trend known to be false, would indeed get one jailed if one were selling it as stock advice.

But that’s not what was going on here. The questions are more subtle, and demanding more accuracy. The chart was to show the trends of global temperatures over a long period of time, with projections past the current date.

I’m sure I won’t convince you based on what you said above, but every Boy Scout and former Scout will understand: Substituting the figures from what actually happened was the right thing to do. It showed what actually happened, and cutting off the dendro proxy data prevented the presentation of a much more erroneous projection to the future.

As you know, the actual temperature readings after the chart’s current date were higher than projected. Had the erroneous data you ask to be used been used, the chart would have been even father wrong. The panic that might have resulted from such an error is incalculable.

Ethics or not, they still felt the compelling need to “hide the decline” by substituting one set of data for another, and that’s wrong.

What they did stands the 4-Way test of Rotarians. What they did is entirely within the Scout Law.

I’m shocked to see Anthony Watts say it is “wrong” to delete information known to be inaccurate and substitute much more accurate information. That’s only “wrong” if one is cheering for total disaster.

But on an ethical scale, what they did stands near the top: Getting good information out is alwasy more valuable than telling falsehoods. And make no doubt about it, temperatures did not begin an precipitous decline in 1960.

It’s like telling the US population that the dollar is strong by splicing the recent values of the Yen onto the value of the dollar data after it started declining and showing it all in one graph, saying “see, the dollar is robust!”. That doesn’t fly, and neither should mixing proxy and real data.

I’d really love to see you justify using known-to-be-incorrect data in place of actual temperature readings. On one hand you campaign for perfectly accurate placement of the devices that measure temperatures, and here you claim that use of that accurate data is “dishonest.” You’re closer to right the first time.

It’s just plain wrong and was only done to strengthen their point when the data nature provided went the way they didn’t want it to.

As it was, the chart showed less warming than actually occurred in the decade or so that has elapsed since the chart was published. I’m not sure what making a chart even more wrong on your side of the scale would have done, Anthony. Think about it for a moment: What would the headlines have read had people said “warming to hit plateau for a decade or so” in 1995, and then we’d gone through 1998 and the well-above projected temperatures from 2000 to 2009. The headlines would have yelled about how much more warming was occurring than the scientists had predicted.

It would be as if the television weatherman predicted warm, sunny weekends for the entire summer, and each weekend was cold and rainy. How long would that weatherman last at the station?

Even for the skeptical side — especially for the skeptical side — those guys did everyone a favor when they strove for accuracy over a foolish consistency.

Accuracy is to be prized above politically-motivated data interpretations, in science.

REPLY: I’m now convinced that your view of reality is hopelessly distorted. Say whatever you like, as it has no impact. But the fact remains that a portion of the tree ring dataset that showed a decline in temperature post 1961 was combined with data that showed the opposite result. If a stock report did something like this for a major corporation stock trend, people would go to jail for defrauding investors.

To help you understand, please see this:

- Anthony

Seems station-data is important now;

Climategate på finsk TV gir en bra oversikt

kwik skrev for 0 minutter siden:

Vi ser ikke stort om Climategate på norsk TV.

De prøver nok å holde lokket på så lenge som mulig.

Men her er 3 deler på Youtube fra en finsk film om Climategate;

There are relatively accurate actual measures for temperature and CO2 going back to the 1800s before that there are historical records and such things as ships logs that can give relatively reliable wind records but once you are back to say the 1600′s then you are left with proxies.

The ENTIRE AGW hypothesis is that temperatures are rising at an exceptional rate and CO2 is at an exceptional level. To do this the it is necessary to know the normal so one can claim the current climate behavior is exceptional. The ‘normal’ climate behavior is only available through proxies and some historic records. For the AGW hypothesis to be accepted the proxies needed to show that the known warm periods such as the Medieval Warm Period and the Roman Optimum were not actually as warm as today. This I believe is what drove the proxy research. Even historic ‘global’ CO2 is based on ice core data the problem is (as the satellite imagery shows) that the CO2 concentration at the poles is significantly lower than elsewhere on the Earth and CO2 also diffuses through ice over time this results in an unnaturally low ‘historic global CO2 from ice cores’ it is unsurprising that historic ice core records show low CO2. What IS surprising is the disregard of actual measures of CO2 taken in the 1800s that show far higher CO2 concentrations.

Climate research is not weather research it is based on multi-century timescales. We have the ability to measure things from satellite now that is only 40 years old at most. But climatology studies this entire inter-glacial and other glacial and inter-glacials and have to rely n proxies. It is time that an independent validation exercise/audit was carried out on all these proxies as peer review is insufficient. A proxy that departs from actual measures in current times may well have similarly departed in previous times and is therefore unsafe.

Anthony has answered the issue of concealing proxies that fail validation.

If you read the CRU emails you will see that they _were_ told that there were far too many variables affecting tree growth rate for it just to be temperature. For example a hot dry year will result in less growth than a temperate wet year, a pest or fungus infestation will result in less growth. This is not ‘noise’ it is that many variables affect tree growth and use for just temperature is unsafe. Especially when the claims are for accuracy to better than plus or minus a few degrees

And if you read the papers you’ll see that those problems were taken into consideration. No one argues that tree rings substitute for thermometer readings, but as a tool to corroborate other measures, they work just fine. The original allegation here was that the dendrochronologists don’t know botany. That’s demonstrably wrong, and the papers do consider the stuff you claim they don’t consider. Read the papers.

If you must use a proxy for a variable then your research and results should also include totally open validation of that proxy to prove that it does act as a reliable proxy.

The research is totally open. Read the papers.

You’re complaining about a chart based off the research papers. The chart was designed to show how dendrochronological data correlate and corroborate other data dealing with global warming. As with many other measures in many other areas — carbon dating has to deal with increased carbon after about 1850, too, for one example — there are variations in the correlation of the proxy measure over time. These variations are well known, not secret, and in this case have been subject to more than a decade of rather vigorous public debate about how to deal with them.

Anyone who reads the literature knows that the dendro data depart from the general trend in the 1950s or 1960s. We don’t know why — it may be, as with the carbon dating data, that the increased amount of carbon has something to do with it. Or it may be that acid rain predominates the effects shown in dendro data at that time, which is about the same time acid rains started to become a major factor. There is a wide variety of potential causes. More research may tease out which is the ultimate cause, or which are the causes.

But in any case, we know that the tree-ring data stop being a good proxy after about 1960. So, for the purposes of this one chart, real temperature readings were substituted.

In other words, known-to-be-accurate data were substituted for known-to-be-inaccurate data.

I’d love to hear one of the warming skeptics explain why actual data should be considered “fraud” in this case. In the real world, that’s a good move, to put in more accurate data in a chart. In any case, this was the substitute data.

If the proxy data are not accurate for the years prior to 1960, the inaccuracies should be explained in the discussions in the papers prior to 1998. Can you point me to a paper that says the data don’t work at all? I’ve not found it. In general it’s fair to say that the proxy data are well known to be well correlative to temperature readings in places and times we couldn’t have temperature readings, with the qualifications well identified in the papers.

This was not done.

Seriously? What do you call the debate over the previous decade? Your argument that we should have used inaccurate data is specious from the start, and here you claim that the difficulties with the proxy data were not known — when your only sources of information on the difficulties are the research papers already on the record identifying those problems.

This shouldn’t be so convoluted. Are the data inaccurate? Can you show us where and how?

Indeed when the unreliability of the proxy became apparent – everyone involved – agreed to hide it.

That’s a false statement on your part. The unreliability of the proxy applied only after 1960, and it’s well discussed in the papers. None of those papers was pulled. All of them remain in public, in research journals.

Your failure to study the issue is not “hiding the data” on the part of the people who did the work.

Now you may have different ethics to the rest of the scientific community – but most scientists report their results openly even if they are unfavorable to their research. Research is intended to increase knowledge, not prove preconceptions.

You may not subscribe to science ethics which requires an accurate summary of the data one criticizes. But that doesn’t affect he accuracy of the overall data. It only means that trying to figure out what you’re saying will be difficult for laymen — you’ve hidden the facts. Odd, isn’t it? You’re doing what you falsely accuse the scientists of doing, and you’re crowing that your ethics are superior.

Research can increase knowledge only if people read it.

so Ed, climate research is based on proxies.

That’s false. Among the more famous non-proxy research, climate research is based on measurements of CO2 on the mountains of Hawaii over the past 60 years, on temperature readings from various science agencies kept over the last 400 years, on the coming of spring, on the zones in which plants grow, on the first freezes and last freezes of a year, on the extent of glaciers as measured during the past 300 years, and a variety of other non-proxy data.

If you were laboring under the misconception that climate data is proxy data, or mostly proxy data, or significantly proxy data, we’ve identified a major source of error for your other observations.

Do you have a problem with validating these proxies as correct and reporting the validation carried out their level of accuracy or inaccuracy, and any ‘adjustments’ as part of that research?

I have no difficulty in using proxies for the purposes they are intended, so far as they tend towards accuracy we can use. In this case, dendrochronology data are used to substitute for actual temperature measurements in times and places we don’t have actual measurements. The proxies are calibrated against real measurements. Adjustments for accuracy are par for the course, to be expected, and not to be fought tooth and nail by people who wish for accuracy.

What is your real question, I cannot tell. If you’re asking whether the proxy data corroborate other measures that show super-warming, the answer is yes. If you’re asking whether proxy data drive climate research, the answer is no.

I ask as it is notable that the IPCC committee (at the meeting in Tanzania) and University of East Anglia and Penn State University appear to be of the opinion that you should hide proxy validation failures and original data.

That’s a false claim.

Perhaps scientific ethics have changed?

Nope, still much stiffer than blog ethics.

REPLY: Ethics or not, they still felt the compelling need to “hide the decline” by substituting one set of data for another, and that’s wrong. It’s like telling the US population that the dollar is strong by splicing the recent values of the Yen onto the value of the dollar data after it started declining and showing it all in one graph, saying “see, the dollar is robust!”. That doesn’t fly, and neither should mixing proxy and real data. It’s just plain wrong and was only done to strengthen their point when the data nature provided went the way they didn’t want it to.

All explained in detail with supporting evidence here: http://climateaudit.org/2009/12/10/ipcc-and-the-trick/

-Anthony

If you must use a proxy for a variable then your research and results should also include totally open validation of that proxy to prove that it does act as a reliable proxy.

This was not done. Indeed when the unreliability of the proxy became apparent – everyone involved – agreed to hide it. Now you may have different ethics to the rest of the scientific community – but most scientists report their results openly even if they are unfavorable to their research. Research is intended to increase knowledge, not prove preconceptions.

so Ed, climate research is based on proxies. Do you have a problem with validating these proxies as correct and reporting the validation carried out their level of accuracy or inaccuracy, and any ‘adjustments’ as part of that research?

I ask as it is notable that the IPCC committee (at the meeting in Tanzania) and University of East Anglia and Penn State University appear to be of the opinion that you should hide proxy validation failures and original data.

Perhaps scientific ethics have changed?

You can’t be bothered to read the papers and the science?

David, how many different sites measured does it take to equal good science on the issue? How do you know? What sort of hubris was it that caused you to think these scientists didn’t bother to get more than one location?

You can’t be bothered to read the papers, either?

This became very apparent when the proxy temperature guesses from the tree rings went down after 1960 when the actual temperatures were going up. This actually showed that the proxies were invalid. If they were invalid after 1960 then they were invalid before 1960 despite possible short term correlations. Thus all the reasoning based on THE tree at Yamal etc was based on invalid assumptions. The entire argument using tree rings was baseless. Instead of being honest about this – ‘the team’ hid this validation failure.

So the validation failure exposed and compounded a larger ethics failure in University of East Anglia.

The problem is misleading policy makers and the public as to the certainty of the reconstructions and the reliability of these tree-ring based temperature reconstructions.

Unbound:

How do you know ” with tree rings that are no longer consistent in size due to the increased carbon emissions in the atmosphere. “?

Your statement quoted is just a made up arm waving dismissal of data to reach a preconceived conclusion. There is no way to know that long term treemometers ever existed in the first place, except for some apparently cherry picked data correlations.

The problem is how to deal with tree rings that are no longer consistent in size due to the increased carbon emissions in the atmosphere. The “trick” to account for the messed up tree rings after 1960 so they could see if trending after correction to the tree rings would continue to line up with the recorded temperatures.

Don’t act like you are trying to find the truth if you are not actually going to look for it.