Climate Science: Is it currently designed to answer questions?

[lomiller]

The Little Ice Age and Medieval Warm Period in the Sargasso Sea
(Science, Vol. 274. no. 5292, pp. 1503 - 1508, November 29, 1996)
- Lloyd D. Keigwin

The Little Ice Age and Medieval Warming in South Africa
(South African Journal of Science 96: 121-126, 2000)
- P. D. Tyson, W. Karlén, K. Holmgren and G. A. Heiss

The Little Ice Age as Recorded in the Stratigraphy of the Tropical Quelccaya Ice Cap
(Science, Vol. 234. no. 4774, pp. 361-364, October 17, 1986)
- L.G. Thompson, E. Mosley-Thompson, W. Dansgaard, P.M. Grootes

I find it interesting that the best you can do is to cite Lonnie Thompson who clearly doesn’t support anything you are saying.

In any case, how can two outdated *regional* studies be empirical proof of *global* warming? (The third isn’t a high impact journal, so I won’t bother to address it)

 

[lomiller]

It is a bit hypocritical that when skeptics point out that the warmers are using to short of a time span it is ignored, but if any skeptics use to short of a time span it is used as an argument against them.

There is no trend calculation involved when you look at record years. Trend calculations require sufficient time to account for natural variability. record years by definition can't take more then a year...

 

[lomiller]

Any measurement without a solid error estimation is without scientific merit.

which is of course why all the major temperature data sets calculate error bands.

 

[a_unique_person]

My argument was never about our understanding of the environment but whether the REAL world is 100% REAL or not, it is - virtual reality is not.

And a very strange argument it is, too. I can't understand what the point of it is. If Poptech has a pointless argument about nothing, does anyone care?

 

[Poptech]

I find it interesting that the best you can do is to cite Lonnie Thompson who clearly doesn’t support anything you are saying.

The paper clearly supports that the LIA age existed globally.

The fact that the Little Ice Age (about A.D. 1500 to 1900) stands out as a significant climatic event in the oxygen isotope and electrical conductivity records confirms the worldwide character of this event.

I love how if a scientist supports AGW alarmism you can never cite their work on anything except in support of AGW alarmism.

In any case, how can two outdated *regional* studies be empirical proof of *global* warming? (The third isn’t a high impact journal, so I won’t bother to address it)

Thankfully you do not get to define what is "outdated" or "high impact". You failed horribly to scientifically prove what "high impact" means. Your subjective nonsense is getting old. These papers are clear evidence of the Little Ice Age that you deny. I am well aware AGW alarmists are all little ice age deniers.

 

[Poptech]

And a very strange argument it is, too. I can't understand what the point of it is. If Poptech has a pointless argument about nothing, does anyone care?

Yes, of course the argument between the the REAL and Virtual (NOT-REAL) world is strange and you cannot understand it - sure. Thankfully many people who are not pushing AGW alarmism can CLEARLY understand it.

 

[MattusMaximus]

My argument was never about our understanding of the environment but whether the REAL world is 100% REAL or not, it is - virtual reality is not.

Actually, your original argument was about accuracy, and you claimed that computer climate models couldn't be trusted because they aren't 100% accurate in their predictive capabilities.

In fact, here are your statements to that effect...

Linky #1

My dismissal of the use of computer climate models for scientific conclusions, predictions and policy is based solely on computer science. The models are not a 100% perfect reproduction of the earth and all it contains in real time, therefore they cannot predict anything. I am just amazed at the lack of understanding people have with computer systems.

Linky #2

100% accuracy is required for relevant results. But weather models are tracking real time data and essentially just extrapolating basic weather movement, cloud cover, rain ect... Long term weather modeling is where it becomes laughable. Climate models are attempting to simulate the climate and give temperature projections 100 years from now. Total nonsense.

And I pinned you down on these arguments, because you kept going on about how we must rely only upon empirical experimentation and it was pointed out to you that not even empirical experimentation would ever be 100% accurate either. So if that's true, then why do you cast stones at a less than 100% computer modeling method but not at less than 100% accurate empirical experiments? You must address this logical inconsistency within your arguments, but to this point you refuse to do so.

Instead, now you are claiming that you never were discussing this, and you keep blathering on and on about the "100% REAL world" and similar nonsense.

Keep on running away, Poptech. You are destroying your credibility by acting in this manner. And because we can track the discussion, everyone here can see it. I'll make certain they do.

 

[Uncayimmy]

Instead, now you are claiming that you never were discussing this, and you keep blathering on and on about the "100% REAL world" and similar nonsense.

And he avoided the very simple question I asked: Without models, how do you plan for the future?

 

[Poptech]

And I pinned you down on these arguments, because you kept going on about how we must rely only upon empirical experimentation and it was pointed out to you that not even empirical experimentation would ever be 100% accurate either. So if that's true, then why do you cast stones at a less than 100% computer modeling method but not at less than 100% accurate empirical experiments? You must address this logical inconsistency within your arguments, but to this point you refuse to do so.

You only pinned something down in your own confused mind. You are confusing two separate issues, the accuracy of the "laboratory" and the "experiment". There is no logical inconsistency just your repeated confusion to understand the difference. For "laboratories" only the REAL world is 100% accurate and thus experiments run in it do not have this bias. Virtual World "experiments" fail the very basics of having a 100% accurate "laboratory" to perform them in. Thus logically it can be concluded only REAL world (empirical) experiments have any remote possibility of being 100% accurate.

Please make certain your continued failure to understand the difference between the REAL and Virtual (NOT-REAL) world is pointed out. Computer illiteracy remains strong.

 

[Poptech]

Without models, how do you plan for the future?

There is no such thing as an accurate predictive computer model for the climate or economics. So you cannot plan based on them, sorry to break it to you. People with control problems might go crazy which is why they refuse to accept it.

 

[Uncayimmy]

There is no such thing as an accurate predictive computer model for the climate or economics. So you cannot plan based on them, sorry to break it to you. People with control problems might go crazy which is why they refuse to accept it.

You said that all computer models were fatally flawed and could predict nothing. Are you amending that statement to restrict it to economic and climate models?

Beyond that, define accuracy. Are there climate models that predicted the mean global temperature in 2008 to drop 50 degrees Celsius? Did any of the economic models predict the Dow to reach 45,000 in 2008? Have any of them made such dramatically inaccurate predictions?

If not, then it would stand to reason that there is a level of accuracy involved. You say that they can predict nothing. Define nothing. If a model predicts (just an example out of my ass) an 80% chance 0.3 degree rise over a decade and there's actually a 0.2 degree rise, what does that mean?

Essentially, what I am asking for is your philosophy behind making the distinction between "inaccurate" and "an acceptable lack of precision." After all, a prediction of something is also an implicit prediction of a whole bunch of not something else.

Local weather forecasts are a great example. Here in Phoenix they are calling for it to reach 110 tomorrow. I am very confident it will be around 108 to 112 at the highest. I would be willing to bet my house that it's not going to reach 125 or drop down to 50. If it reaches 113, I would call that a lack of precision, not unreliably inaccurate.

 

[Poptech]

You said that all computer models were fatally flawed and could predict nothing. Are you amending that statement to restrict it to economic and climate models?

Show me where I have ever talked about anything else?

Beyond that, define accuracy. Are there climate models that predicted the mean global temperature in 2008 to drop 50 degrees Celsius? Did any of the economic models predict the Dow to reach 45,000 in 2008? Have any of them made such dramatically inaccurate predictions?

Close enough is not right on a computer it is wrong. 1.5 is not a close enough answer to 1+1, it is wrong. You could extrapolate historic temperatures in excel and come "close enough". Enough people could guess and someone will be "close enough". Did any of the economic models predict the DOW at 6500 in March of 2009 years in advance? Nope.

If you don't understand the limitations of computers then you want to believe there is a level of accuracy in the predictions. In reality they are no better than guessing and actually worse because "scientific authority" is attached to them all due to computer illiteracy.

I only have a problem with computer climate models when they are used for predictions, scientific conclusions and public policy.

Local weather forecasts are a great example. Here in Phoenix they are calling for it to reach 110 tomorrow. I am very confident it will be around 108 to 112 at the highest. I would be willing to bet my house that it's not going to reach 125 or drop down to 50. If it reaches 113, I would call that a lack of precision, not unreliably inaccurate.

I thought weather has nothing to do with climate?

 

[Herzblut]

I only have a problem with computer climate models when they are used for predictions, scientific conclusions and public policy.

That's a wise notion because climate cannot be predicted. Nobody (to be taken serious) ever claimed climate to be predictable.

You haven't figured that the IPCC et al are talking about "projections", not predictions?

 

[Poptech]

The IPCC report makes scientific conclusions based on computer climate models that public policy is now being based.

Projection (Defined) - "A prediction or an estimate of something in the future, based on present data or trends."

It is the same thing.

 

[Uncayimmy]

Show me where I have ever talked about anything else?

There would be no point in that if you are now clarifying that by "computer model" you mean "computer model used to project climate changes."

Close enough is not right on a computer it is wrong. 1.5 is not a close enough answer to 1+1, it is wrong.

I agree that 1.5 is not an acceptable answer to 1 + 1. However, I believe it is a perfectly acceptable answer to 2.4999999999 - 1.000001 under certain circumstances. It depends on what I'm doing it.

Have you ever programmed computers? We deal with issues of precision in floating point numbers on a regular basis. I'm not going to embark on a lecture about different data types, but suffice it to say programmers have to make choices about precision when designing a system.

If you want to insist that "close enough" is always wrong, that's fine by me. It doesn't change the fact that close enough is still close enough and extremely useful.

You could extrapolate historic temperatures in excel and come "close enough". Enough people could guess and someone will be "close enough".

The people extrapolating numbers will be closer far more often than people guessing. The question you are avoiding is where do you personally transition from wrong to within tolerances?

Did any of the economic models predict the DOW at 6500 in March of 2009 years in advance? Nope.

Economic models are another subject entirely. I will not debate that with you.

If you don't understand the limitations of computers then you want to believe there is a level of accuracy in the predictions. In reality they are no better than guessing and actually worse because "scientific authority" is attached to them all due to computer illiteracy.

Define guessing. Differentiate between my three year old "guessing" and a group of 10 scientists "guessing" the mean global temperature in the year 2020.

I only have a problem with computer climate models when they are used for predictions, scientific conclusions and public policy.

So, once again, how do you plan for the future?

I thought weather has nothing to do with climate?

I used the weather as an example of the issue precision and accuracy. You, of course, knew that. You're just being deliberately obtuse.

When the weatherman says it will reach 110 tomorrow, he is by definition excluding a huge range of other possible values. If the temp reaches 111, he was "wrong" as far as you're concerned. Is he just as wrong if he predicted it would only reach 37?

Just as an example of your train of thought, explain to me what tolerances you would accept for temperature predictions for the next day. I'll go first. I figure if he's within +/-5 degrees Fahrenheit 95% of the time, I will find his predictions useful and reliable. What about you?

 

[BenBurch]

My commodities trader friend is much richer because HIS model predicted both the market crash and the bottom at 6600... It was two months off on when the bottom was though.

 

[Megalodon]

Interesting data mining Megalodon, but all the graphs I have seen have shown 1998 to be the highest year in recent history.

Your ignorance is your own problem. All my graphs use the NCDC data, and are accurate representations of the dataset. Your accusations of data mining require evidence (which you don't have), otherwise you are a stupid liar (painfully obvious).

It is a bit hypocritical that when skeptics point out that the warmers are using to short of a time span it is ignored, but if any skeptics use to short of a time span it is used as an argument against them.

I could ask you what is it that you are trying so hard and unsuccessfully to say, but I do doubt it would interest me...

 

[Megalodon]

It is not unprecented.

I don't care about links, specially the ones from infantile posters who constantly misunderstand what's written in them. The data is out there, show me where there is an historical precedent.

And Antarctica is not melting.

Even if you were right (you aren't) Antarctica is a part of the global cryosphere, so it would be irrelevant, unless it was freezing in a way to offset the rest of the world.

Which is not a problem.

From your link: "It is also calculated –using an equilibrium model of aqueous carbonate species in seawater open to the atmosphere-"

Can't have it both ways. Either models work or they don't.

 

[a_unique_person]

Yes, of course the argument between the the REAL and Virtual (NOT-REAL) world is strange and you cannot understand it - sure. Thankfully many people who are not pushing AGW alarmism can CLEARLY understand it.

You don't understand. Every equation we use in science is a model.

 

[Uncayimmy]

You don't understand. Every equation we use in science is a model.

I think I get the gist of his argument. He is bascially stating that it is impossible to create a climate model that is 100% accurate. I agree. Even if we exclude unforeseen events (a catastrophic volcanic eruption would screw up projections immensely), there are just too many variables and not enough "knowledge" of the interactions to say with 100% certainty what the climate will be like in N years.

What he's doing is making it a black and white issue. If the models cannot be 100% accurate, then we are not entitled to use them in any fashion whatsoever. And since the "experts" seem (to him) to be treating these models as completely accurate when they are not, all of their conclusions must be wrong. Furthermore, they must be fools for not recognizing this or they have an agenda and are deliberately ignoring the "accuracy problems" and lying to the public.

I have created my own computer model of Poptech. I predict with 99.99% certainty that he will not be persuaded otherwise.