[AJS]

Are you saying that models make incorrect predictions more often or just as often as they make correct ones?

No.

I have no idea what the ratio is. I am saying exactly what I said. They can't be relied on to make consistently accurate predictions, and we should take this uncertainty into account when forming policy to respond to it.

[Pants-of-dog]

If you don't know what the ratio is, then it might be 100% accurate, or 95% accurate.

This would then suggest that your argument is wrong, and that they can be relied on.

[AJS]

But it might also be 50% or 5% or 0%. The reality is nobody knows how accurate these models are.

[Drlee]

But it might also be 50% or 5% or 0%. The reality is nobody knows how accurate these models are.

This is not true. We know with some certainty that the science in most of the man-made global climate change is sound.

Here is the thing AJS. The data are overwhelmingly convincing.

[Pants-of-dog]

But it might also be 50% or 5% or 0%. The reality is nobody knows how accurate these models are.

Actually, you just said you know how accurate these models are. You were just vague about the numbers.

This is actually a logical fallacy called argument from ignorance.

You are saying that you do not know how accurate the models are, therefore the models cannot be accurate to the point of making good predictions.

But if you do not know how accurate the models are, they could be just fine.

This logical fallacy is often used to shift the burden of proof. If I were to ask you to support your claim (i.e. that climate models cannot consistently make accurate predictions), you will say that it is somehow impossible to prove this and tell me to prove that they are accurate predictors.

So, do you have any evidence for your claim?

[Drlee]

The thing is POD, that not very smart people are driving a debate where no debate ought to exist. We are asked to believe that a vast, almost universal, conspiracy exists between virtually all scientists, in virtually every discipline, to somehow pervert the science. And to what end? What do they stand to gain? Nothing. Nothing at all.

I know that most conservatives are inclined to believe that NASA lies but here is what they said:

Multiple studies published in peer-reviewed scientific journals1 show that 97 percent or more of actively publishing climate scientists agree: Climate-warming trends over the past century are extremely likely due to human activities. In addition, most of the leading scientific organizations worldwide have issued public statements endorsing this position. The following is a partial list of these organizations, along with links to their published statements and a selection of related resources.

Among these are the National Academy of Science:

The scientific understanding of climate change is now sufficiently clear to justify taking steps to reduce the amount of greenhouse gases in the atmosphere." (2005)11

The National Geophysical Union:

"Human‐induced climate change requires urgent action. Humanity is the major influence on the global climate change observed over the past 50 years. Rapid societal responses can significantly lessen negative outcomes." (Adopted 2003, revised and reaffirmed 2007, 2012, 2013)

The National Meteorological Society:

"It is clear from extensive scientific evidence that the dominant cause of the rapid change in climate of the past half century is human-induced increases in the amount of atmospheric greenhouse gases, including carbon dioxide (CO2), chlorofluorocarbons, methane, and nitrous oxide."

The American Association for the Advancement of Science:

"The scientific evidence is clear: global climate change caused by human activities is occurring now, and it is a growing threat to society."

Then there are these players:


Academia Chilena de Ciencias, Chile
Academia das Ciencias de Lisboa, Portugal
Academia de Ciencias de la República Dominicana
Academia de Ciencias Físicas, Matemáticas y Naturales de Venezuela
Academia de Ciencias Medicas, Fisicas y Naturales de Guatemala
Academia Mexicana de Ciencias,Mexico
Academia Nacional de Ciencias de Bolivia
Academia Nacional de Ciencias del Peru
Académie des Sciences et Techniques du Sénégal
Académie des Sciences, France
Academies of Arts, Humanities and Sciences of Canada
Academy of Athens
Academy of Science of Mozambique
Academy of Science of South Africa
Academy of Sciences for the Developing World (TWAS)
Academy of Sciences Malaysia
Academy of Sciences of Moldova
Academy of Sciences of the Czech Republic
Academy of Sciences of the Islamic Republic of Iran
Academy of Scientific Research and Technology, Egypt
Academy of the Royal Society of New Zealand
Accademia Nazionale dei Lincei, Italy
Africa Centre for Climate and Earth Systems Science
African Academy of Sciences
Albanian Academy of Sciences
Amazon Environmental Research Institute
American Academy of Pediatrics
American Anthropological Association
American Association for the Advancement of Science
American Association of State Climatologists (AASC)
American Association of Wildlife Veterinarians
American Astronomical Society
American Chemical Society
American College of Preventive Medicine
American Fisheries Society
American Geophysical Union
American Institute of Biological Sciences
American Institute of Physics
American Meteorological Society
American Physical Society
American Public Health Association
American Quaternary Association
American Society for Microbiology
American Society of Agronomy
American Society of Civil Engineers
American Society of Plant Biologists
American Statistical Association
Association of Ecosystem Research Centers
Australian Academy of Science
Australian Bureau of Meteorology
Australian Coral Reef Society
Australian Institute of Marine Science
Australian Institute of Physics
Australian Marine Sciences Association
Australian Medical Association
Australian Meteorological and Oceanographic Society
Bangladesh Academy of Sciences
Botanical Society of America
Brazilian Academy of Sciences
British Antarctic Survey
Bulgarian Academy of Sciences
California Academy of Sciences
Cameroon Academy of Sciences
Canadian Association of Physicists
Canadian Foundation for Climate and Atmospheric Sciences
Canadian Geophysical Union
Canadian Meteorological and Oceanographic Society
Canadian Society of Soil Science
Canadian Society of Zoologists
Caribbean Academy of Sciences views
Center for International Forestry Research
Chinese Academy of Sciences
Colombian Academy of Exact, Physical and Natural Sciences
Commonwealth Scientific and Industrial Research Organization (CSIRO) (Australia)
Consultative Group on International Agricultural Research
Croatian Academy of Arts and Sciences
Crop Science Society of America
Cuban Academy of Sciences
Delegation of the Finnish Academies of Science and Letters
Ecological Society of America
Ecological Society of Australia
Environmental Protection Agency
European Academy of Sciences and Arts
European Federation of Geologists
European Geosciences Union
European Physical Society
European Science Foundation
Federation of American Scientists
French Academy of Sciences
Geological Society of America
Geological Society of Australia
Geological Society of London
Georgian Academy of Sciences
German Academy of Natural Scientists Leopoldina
Ghana Academy of Arts and Sciences
Indian National Science Academy
Indonesian Academy of Sciences
Institute of Ecology and Environmental Management
Institute of Marine Engineering, Science and Technology
Institute of Professional Engineers New Zealand
Institution of Mechanical Engineers, UK
InterAcademy Council
International Alliance of Research Universities
International Arctic Science Committee
International Association for Great Lakes Research
International Council for Science
International Council of Academies of Engineering and Technological Sciences
International Research Institute for Climate and Society
International Union for Quaternary Research
International Union of Geodesy and Geophysics
International Union of Pure and Applied Physics
Islamic World Academy of Sciences
Israel Academy of Sciences and Humanities
Kenya National Academy of Sciences
Korean Academy of Science and Technology
Kosovo Academy of Sciences and Arts
l'Académie des Sciences et Techniques du Sénégal
Latin American Academy of Sciences
Latvian Academy of Sciences
Lithuanian Academy of Sciences
Madagascar National Academy of Arts, Letters, and Sciences
Mauritius Academy of Science and Technology
Montenegrin Academy of Sciences and Arts
National Academy of Exact, Physical and Natural Sciences, Argentina
National Academy of Sciences of Armenia
National Academy of Sciences of the Kyrgyz Republic
National Academy of Sciences, Sri Lanka
National Academy of Sciences, United States of America
National Aeronautics and Space Administration
National Association of Geoscience Teachers
National Association of State Foresters
National Center for Atmospheric Research
National Council of Engineers Australia
National Institute of Water & Atmospheric Research, New Zealand
National Oceanic and Atmospheric Administration
National Research Council
National Science Foundation
Natural England
Natural Environment Research Council, UK
Natural Science Collections Alliance
Network of African Science Academies
New York Academy of Sciences
Nicaraguan Academy of Sciences
Nigerian Academy of Sciences
Norwegian Academy of Sciences and Letters
Oklahoma Climatological Survey
Organization of Biological Field Stations
Pakistan Academy of Sciences
Palestine Academy for Science and Technology
Pew Center on Global Climate Change
Polish Academy of Sciences
Romanian Academy
Royal Academies for Science and the Arts of Belgium
Royal Academy of Exact, Physical and Natural Sciences of Spain
Royal Astronomical Society, UK
Royal Danish Academy of Sciences and Letters
Royal Irish Academy
Royal Meteorological Society (UK)
Royal Netherlands Academy of Arts and Sciences
Royal Netherlands Institute for Sea Research
Royal Scientific Society of Jordan
Royal Society of Canada
Royal Society of Chemistry, UK
Royal Society of the United Kingdom
Royal Swedish Academy of Sciences
Russian Academy of Sciences
Science and Technology, Australia
Science Council of Japan
Scientific Committee on Antarctic Research
Scientific Committee on Solar-Terrestrial Physics
Scripps Institution of Oceanography
Serbian Academy of Sciences and Arts
Slovak Academy of Sciences
Slovenian Academy of Sciences and Arts
Society for Ecological Restoration International
Society for Industrial and Applied Mathematics
Society of American Foresters
Society of Biology (UK)
Society of Systematic Biologists
Soil Science Society of America
Sudan Academy of Sciences
Sudanese National Academy of Science
Tanzania Academy of Sciences
The Wildlife Society (international)
Turkish Academy of Sciences
Uganda National Academy of Sciences
Union of German Academies of Sciences and Humanities
United Nations Intergovernmental Panel on Climate Change
University Corporation for Atmospheric Research
Woods Hole Oceanographic Institution
Woods Hole Research Center
World Association of Zoos and Aquariums
World Federation of Public Health Associations
World Forestry Congress
World Health Organization
World Meteorological Organization
Zambia Academy of Sciences
Zimbabwe Academy of Sciences

Only a person of monumental stupidity would even think about questioning the fact of climate changed caused by people and their use of fossil fuel. I can't stress this enough. When I see someone here even questioning the data it is clear to me that we are dealing either with someone of low intelligence coupled with poor education or someone who is deliberately trying to exploit those just mentioned. If this is you I can unequivocally assert that your taking the test for Mensa will be a waste of money which would better be spent on getting your GED.

If it sounds like I am angry about this, I am. These piss ants are selling out their children and grand children for a few bucks today. In that way they are very close to advocating mass murder for the profit of a few large corporations. They are the lowest of the low.

[AJS]

Actually, you just said you know how accurate these models are. You were just vague about the numbers.

This is actually a logical fallacy called argument from ignorance.

You are saying that you do not know how accurate the models are, therefore the models cannot be accurate to the point of making good predictions.

But if you do not know how accurate the models are, they could be just fine.

This logical fallacy is often used to shift the burden of proof. If I were to ask you to support your claim (i.e. that climate models cannot consistently make accurate predictions), you will say that it is somehow impossible to prove this and tell me to prove that they are accurate predictors.

So, do you have any evidence for your claim?

No, I said I know the limitations of those models. That's not the same as saying I know precisely how often they will be correct.

[Drlee]

No, I said I know the limitations of those models. That's not the same as saying I know precisely how often they will be correct.

Read my post above. You "know" no such thing. If you did you would not be having this silly conversation.

[Perkwunos]

@Pants-of-dog

Yes I know what climate models are and I know the statistical/mathematical tools used for those models. I also know te limitations of such models in definitively establishing causal relationships or accurately predicting future changes in complex systems.

Any such model is only as good as the assumptions underpinning it and the information fed into it.

In the field of economics no models yet developed have accurately and consistently predicted share price movements, currency fluctuations, GDP growth or commodity prices. Even on a quarterly basis, let alone decades hence. And people have been trying to predict these things for centuries, and putting their own money on those predictions.

The climate of the planet is far more complex and unpredictable than any of these things. Even long range weather forecasts are fairly hit and miss.

At best these models provide and educated guess at what might happen.

An interesting argument, all but for the niggling little detail that the climate models have been largely accurate:

https://www.wunderground.com/climate/fa ... liable.asp

Do you think they just got this right by chance?

Lastly, and this is my speculation, but look at the response. It has in many ways picked up where religion and communism left off. The rich are guilty by implication. The private and individual (especially the car) are bad, while the public and communal are good (even empty buses). A clergy of scientists, consultants and politicians hold the key to our salvation. Some big companies profit to an extraordinary degree in a subsidised market, rich land owners reap huge rewards for wind farms and governments are deemed responsible to the extent that they tax fossil fuels, which remain the driving force of economic activity and will do for decades to come.

There you go interpreting support for the theory entirely in terms of economic self-interest. Do you really think postdocs are in it for the money? Really?

[Pants-of-dog]

No, I said I know the limitations of those models. That's not the same as saying I know precisely how often they will be correct.

Please provide evidence for your claim that climate models do not make consistently correct predictions. Thanks.

[Rapperson]

@AJS

Some observations:

Assumptions this and gaps that - this all amounts to nothing more than handwaving. You can't make this case unless you can quantify what you are talking about. How much of a gap is a problem? How accurate does a particular assumption need to be?

Since you mention it, I'd not be putting my money into safer nuclear stations, what with only 200 years' worth of Uranium left. Doesn't seem like a sound investment to me. Today, 20% of electricity comes from nuclear power. If you want to up that fraction, the candle that burns twice as bright and that. Most people expect solar power is the future, and they are probably right. The price is still too high.

Now, what politicians and activists make of this whole climate conundrum is a story unto itself, and conflating that issue with climate science is highly inappropriate. Very possibly, the issue has been hijacked for nefarious purposes. But that has nothing to do with the science behind it.

[AJS]

@Perkwunos

It's easy to make accurate models for things that have already happened. Correlation is not causation though.


@Rapperson

You can't quantify what you don't know.

[Pants-of-dog]

@AJS

Please provide evidence for your claim that models cannot provide a consistently correct set of predictions. Thanks!

[AJS]

@AJS

Please provide evidence for your claim that models cannot provide a consistently correct set of predictions. Thanks!

I've already said why I have little confidence in the models. It will take decades to know if they are accurate or not in their predictions.

[Vlerchan]

I think the damage that pushing false climate science does probably far outweighs the damage that pushing of bad science in sociology and psychology does.

Whilst the state of academia with regards to the latter is terrible, I never felt as if policy-makers listened to those people all too much.

@Perkwunos

It's easy to make accurate models for things that have already happened. Correlation is not causation though.

1. The only reason that a model which was backwardly-predictive would fail to hold in predicting into the future would be if the parameters used to calibrate our backwardly-predictive model failed to hold in the future*. Unless you can put forward, though, a plausible argument suggestive of a uncontrolled-for shift occurring in one of our parameters - and one which would result in our backwardly-predictive model providing biased predictions where it comes to the future - I don't see much of a reason to really care about your argument. It is the equivalent of, we shouldn't make predictions about future events because who knows.

2. When it comes to multi-variable analysis, correlation does not equal causation is just a cheap argument. If you can't come up with a plausible source of endogeneity in the model being supplied then you've returned to the equivalent of, who knows.

3a. It's is, on the other hand, more than acceptable to claim that we can never know, 100 percent, whether our model is correct. But the who knows-approach, applied consistently, will disable us as far as any sort of mass decision making is concerned, if not decision making - period. This applies just as much to experiments which are conducted under ideal conditions, too. When it comes to the scientific process though, it's long been recognized that what's effectively nihilism isn't an appropriate response to an argument: if there are issues in the data, or the methods, then by all means, direct us to them. But in resorting to the who knows-approach you are rejecting any sort of capacity for science to make a contribution to human life - you are, in effect, undermining the entire idea of science playing a meaningful role in society.

Perhaps that's the sort of society which you would prefer to live in. But, I would think, that the use of the scientific method to guide policy has resulted in demonstrably better results as compared to polities where such an approach hasn't been pursued.

3b. Nevertheless, professional climate scientists typically state that there is a high degree of confidence in the current models.

4. It is not 'easy to make accurate models for things that have already happened'. In fact, constructing an accurate model of past events requires the same knowledge of about the theoretical underpinnings of change across the time-series as it does to construct a model of future events**.

For a verifiable example, go and examine the last 30 years of research for business-cycle models in macroeconomics, where understanding the determinants has been far from 'easy'.

---

* For example, in macroeconomics, if we constructed a backwardly-predictive model of growth and calibrated the model so that it accounted for a Frisch elasticity of 1.0 but, in the future this would be, instead, 0.75, then our backwardly-predictive model would no be forwardly-predictive.

** We then predict on the basis of certain scenarios. For example, the consequences of climate change gets modeled under a number of emissions and abatement scenarios.

Our predictions might be wrong if we are incorrectly estimating a certain parameter but, as discussed, unless you can provide details as to which one we are incorrectly estimating and/or why it would deviate from past trends then I am not sure why your argument should enter the conversation.

[Rapperson]

Everybody should read Vlerchan's post. Excellent contribution.

[Pants-of-dog]

I've already said why I have little confidence in the models. It will take decades to know if they are accurate or not in their predictions.

Oh, I think I understand now.

You are saying that you have a reason for believing that the climate models cannot deliver consistently good results, but you have no factual evidence to support this belief.

Do you know what hindcasting is?

[Perkwunos]

@Perkwunos

It's easy to make accurate models for things that have already happened. Correlation is not causation though.

Well I guess it's a pity that the models don't predict only training set data well:

https://www.wunderground.com/climate/fa ... liable.asp

[AJS]

Our predictions might be wrong if we are incorrectly estimating a certain parameter but, as discussed, unless you can provide details as to which one we are incorrectly estimating and/or why it would deviate from past trends then I am not sure why your argument should enter the conversation.

This is the nub of it.

I don't have a clue what parameters have been missed or incorrectly estimated, but with something as complicated as the climate of the planet and its relationship with carbon and other emissions it seems quite likely that something is missing. Even if a model appears to work over a short period of time it can still be wrong and can still fail dramatically over longer periods.

[Drlee]

I don't have a clue what parameters have been missed or incorrectly estimated, but with something as complicated as the climate of the planet and its relationship with carbon and other emissions it seems quite likely that something is missing.

You are right. You don't have a clue. What evidence do you have other than you, a non-expert, as a matter of faith in the face of no evidence, conclude that it is "Quite likely" that something is missing and that that something will turn the entire case on its ear.

Do you not see how silly this is?