The fiddling with temperature data is the biggest science scandal ever

[Darryl1958]

People are not fudging the data about obscure factors contributing to the climate.
Scientists are fudging the data about climatge change in general though.

Without reliable scientific data, one opinion is just as good as another.
We cannot carry on analyzing the data as if it has not already been compromised.

 

[LeafByNiggle]

Hans, CO2 (man made or otherwise) has shown itself to fail the scientific method.
Instead of experiencing Scenario A as projected, we are below the idealized Scenario C

No projection of future climate trends can ever be perfect. Even the discrepancy shown on your graph, if true, still does not “fail” the scientific method, or disprove the theory.

 

[sps49]

No projection of future climate trends can ever be perfect. Even the discrepancy shown on your graph, if true, still does not “fail” the scientific method, or disprove the theory.

I think the point was that the curve fit (model) failed; it did not fit the facts. If it does not fit, then one must go back and find out why. That’s the bit that is missing in whatever protocol the IPCC or other entities have for releasing data intended for policy makers.

 

[Ridgerunner]

I think the point was that the curve fit (model) failed; it did not fit the facts. If it does not fit, then one must go back and find out why. That’s the bit that is missing in whatever protocol the IPCC or other entities have for releasing data intended for policy makers.

Unfortunately, the current “policy makers” in the U.S. have already made up their minds to make energy more difficult to obtain and more expensive, based on political considerations alone.

 

[LeafByNiggle]

I think the point was that the curve fit (model) failed; it did not fit the facts. If it does not fit, then one must go back and find out why. That’s the bit that is missing in whatever protocol the IPCC or other entities have for releasing data intended for policy makers.

The extent to which empirical data fits any particular model is not a pass/fail thing. It is not correct to say that the model “failed” in any absolute sense. The most accurate thing you can say about it is that it overestimated atmospheric temperature trends over such-and-such a time period by such-and-such an amount. Then begin a discussion of how relevant those particular discrepancies are to the theory upon which the model is based.

 

[estesbob]

The extent to which empirical data fits any particular model is not a pass/fail thing. It is not correct to say that the model “failed” in any absolute sense. The most accurate thing you can say about it is that it overestimated atmospheric temperature trends over such-and-such a time period by such-and-such an amount. Then begin a discussion of how relevant those particular discrepancies are to the theory upon which the model is based.

Actually every single model failed. Not one single on predicted a 18 year “pause” in global warming. These models were used to demand that we embark on a massive worldwide tax and regulation scheme. You cant now simply dismiss them by saying its not a matter of pass/fail.

 

[LeafByNiggle]

Actually every single model failed. Not one single on predicted a 18 year “pause” in global warming.

I do not consider that a failure.

These models were used to demand that we embark on a massive worldwide tax and regulation scheme. You cant now simply dismiss them by saying its not a matter of pass/fail.

How the models were used has no bearing on the scientific question of how accurate they were.

 

[estesbob]

I do not consider that a failure.

How the models were used has no bearing on the scientific question of how accurate they were.

“Pay no attention to the man behind the curtain…”

 

[baliano]

It’s a basic fact the Earth has undergone several severe climate changes throughout geological history, most before there were any humans to influence the climate. To me it’s pretty egotistical for humans to think they can have any serious effect on something as dynamic and global as Climate change. However, those who can influence global aristocrats have and will continue to reap huge fortunes at the expense of the unwitting public in terms of Carbon Taxes, loss of efficient means of producing power and all the other “solutions” that serve to profit a few and hurt the many.

 

[sps49]

The extent to which empirical data fits any particular model is not a pass/fail thing. It is not correct to say that the model “failed” in any absolute sense. The most accurate thing you can say about it is that it overestimated atmospheric temperature trends over such-and-such a time period by such-and-such an amount. Then begin a discussion of how relevant those particular discrepancies are to the theory upon which the model is based.

Sure it is, especially when you can see the deviation with the unaided eye. It has failed in the absolute sense; it did not match the data. Mechanical and electrical engineers don’t get a pass for failed models, neither does anyone else, especially when they may be used to guide public policy. The relevant discussion is how to fix them, not finding a new combination of weasel words to obscure the failure.

 

[LeafByNiggle]

Sure it is, especially when you can see the deviation with the unaided eye. It has failed in the absolute sense; it did not match the data. Mechanical and electrical engineers don’t get a pass for failed models, neither does anyone else, especially when they may be used to guide public policy. The relevant discussion is how to fix them, not finding a new combination of weasel words to obscure the failure.

OK, just so I understand what you mean by a failed model, please tell me the threshold at which you would say a temperature prediction model fails? In other words, how good would the actual temperature record have to be for you to say that the models did **not **fail? And don’t say “I don’t know, but these these surely did fail.”

 

[Theo520]

It can’t be any worse of a fail. Hansen’s CAGW theory predicted Scenario A yet we achieved below Scenario C

You are denying reality to say it doesn’t matter when theory is proven completely wrong.

No projection of future climate trends can ever be perfect. Even the discrepancy shown on your graph, if true, still does not “fail” the scientific method, or disprove the theory.

 

[sps49]

OK, just so I understand what you mean by a failed model, please tell me the threshold at which you would say a temperature prediction model fails? In other words, how good would the actual temperature record have to be for you to say that the models did **not **fail? And don’t say “I don’t know, but these these surely did fail.”

The model curve does not match the temperature record curve. The model temp rates do not follow the measurements, or match the values. If they wish to make policy decisions, then they should clean it up first. To answer you specific question, it is not simply a threshold, but also shape. Does it fit the measurements? No?, start over.

The validation and verification for these are a expert-subjective judgement. That means the models are not judged against a performance specification like a control system model would be, but by the expert opinion of the development team of its output compared to previous versions and measured data. There are no objective measures for this, the closest you can get is an outwardly looking objective opinion which is really a subjective decision based on invisible and subjective scoring. It is in reality a “that looks about right” process cloaked in an opaque process that is designed to appear objective, but is not.

Again, if they are to be used to make policy decisions, fix them so that they at least fit the measured data. If they are to be used only in research, they can do whatever they wish with them. They still have utility in that they may provide clues to processes that may not be measured directly.

 

[LeafByNiggle]

The model curve does not match the temperature record curve. The model temp rates do not follow the measurements, or match the values. If they wish to make policy decisions, then they should clean it up first. To answer you specific question, it is not simply a threshold, but also shape. Does it fit the measurements? No?, start over.

The validation and verification for these are a expert-subjective judgement. That means the models are not judged against a performance specification like a control system model would be, but by the expert opinion of the development team of its output compared to previous versions and measured data. There are no objective measures for this, the closest you can get is an outwardly looking objective opinion which is really a subjective decision based on invisible and subjective scoring. It is in reality a “that looks about right” process cloaked in an opaque process that is designed to appear objective, but is not.

Again, if they are to be used to make policy decisions, fix them so that they at least fit the measured data. If they are to be used only in research, they can do whatever they wish with them. They still have utility in that they may provide clues to processes that may not be measured directly.

So now can you please answer my question and tell me in objective terms what it sort of temperature record it would take for the AGW model not to fail?

 

[lynnvinc]

Beware of wolves in sheep’s clothing.

And…

Saint Michael the Archangel, defend us in battle, be our protection against the wickedness and snares of the devil. May God rebuke him we humbly pray; and do thou, O Prince of the Heavenly host, by the power of God, cast into hell Satan and all evil spirits who wander through the world seeking the ruin of souls. Amen.

 

[sps49]

So now can you please answer my question and tell me in objective terms what it sort of temperature record it would take for the AGW model not to fail?

The models are not tested in an objective way. There is no objective specification for the models, only comparisons to past predictions and the subjective judgement of the team. Have you looked at the validation and verification process’?

The models are being used to justify public policy, and yet they do not match reality. They are not designed with objective standards, and yet you demand an objective standard by which to measure them. One could surmise that by your reasoning, they should not be made public until objective design standards are used as a specification by which they may be measured. Good luck with that, I don’t think that the model teams would submit to that kind of process control, even though most of the rest of the world does.

 

[LeafByNiggle]

The models are not tested in an objective way.

Of course they are. You do it yourself. Or how else could you contend that the models have all failed unless you have, in your own way, tested them, and found them lacking? You can’t have it both ways. You can’t at once claim the models are untestable and at the same time claim they have failed.

 

[zz912]

Because we can trust our Catholic popes to tell the truth – and we can NOT trust the AGW denialist industry hacks posing as climate scientists, who are paid to lie or are rewarded nicely for lying. Also , the popes take it further by indicating the responsibilities we have about AGW, the moral side of the issue. However, they only spell it out in general terms; it is then up to us to figure out concrete ways to mitigate AGW. They also lead by example, having made the Vatican “carbon neutral.”

Popes have declared that AGW is real and this is a matter of faith?

Does the charism of infallibility extend to climate pronouncements?

 

[zz912]

YOU trust the scientists programming the data, and I do not. If truth was on the side of the AGW believers, there would be no need to alter data. Now, what do you propose should be done to solve this “problem” of earth warming?

Usually more socialism is the proposed solution.

 

[sps49]

Of course they are. You do it yourself. Or how else could you contend that the models have all failed unless you have, in your own way, tested them, and found them lacking? You can’t have it both ways. You can’t at once claim the models are untestable and at the same time claim they have failed.

Of course they are not. Here are a couple papers that touch and verification and validation:

researchgate.net/profile/Ekkehard_Holzbecher/publication/235322171_Remarks_on_Testing_ecological_models_the_meaning_of_validation_(Rykiel_E.J._Ecol._Mod._90_229-2441996)/links/0fcfd510bc44c47f8a000000.pdf

informs-sim.org/wsc07papers/014.pdf

The fundamental problem with using expert subjective judgement as the the validation tool, (while papering over or obscuring verification) is that each expert will have a different opinion. different opinions may lead to different results, and a lack of portability of the model, which makes it difficult if not impossible for independent review. This is not a problem if the model is used for exclusively academic purposes, but is a huge issue if used as an (name removed by moderator)ut for public policy.

Climate models have also ultimately failed as tools for policy makers due to the reasons outlined above. There is one way to salvage the model credibility issue, rigorously apply known verification and validation techniques to the problem. No one said it would be easy.

As for me having it both ways, it is not my problem that the modelers painted themselves into a corner.