How Can You Have “Certainty” When So Many Factors are “Unknown?”

ipcc-forcings

Dr. Curry has an interesting post today regarding the “UnCertainty Monster.” Climate Alarmists speak of levels of “certainty” that are simply unsustainable in uncontrolled experiments and “sciences” that are unsupported by experimentation. Weather forecasters can’t predict the weather 5 days out with any accuracy, yet climate “scientists” speak in terms of 95%+ certainty about changes in the global climate 100 years in the future. They also make claims that one factor out of an infinite number of factors that influence climate is responsible for 100% of the warming over the past 100 years. All those claims are pure nonsense.

To build a valid scientific model, one has to control for as many “exogenous” variables as possible, and have accurate data for all “significant” factors defined in the theory. For instance, a valid experiment to prove the law of Gravity would be to:

  1. Secure a standardized Vacuum tube
  2. Secure standardized 1, 3, 5, 10-ounce ball bearings
  3. Secure a standardized clock
  4. Secure a standardized distance measuring scale and drop platform
  5. Secure a standardized touchpad
  6. Identify various locations around the globe that are at sea level

I would then travel around the globe to these sea level locations, setting up my vacuum tube, dropping the various balls from different heights within the vacuum tube, and measuring the time if takes the balls to fall from the platform to the touchpad. I would then publish my research and encourage others to “reproduce” my findings. Others would then take near identical instruments and go about either validating or rejecting my findings. Because that experiment has been performed countless times, and the results are alway the same (objects fall at 9.8m/sec^2), we give it the title “Law” of Gravity. When you talk of “Laws” you can use the terms of 95% certainty. Climate Change is no Law, at best it is a poorly defined “Theory,” at worse, a politically motivated fraud.

Screen-Shot-2017-03-11-at-3.30.43-AM

Climate “science” isn’t like a real science, it isn’t what is called a “hard” science. Climate “science” is a “soft” science and is closer to social and political “science,” than it is to physics and chemistry. Climate “science” is what happens when political activists in the social science departments invade the turf typically held and protected by the math, physics, chemistry and engineering departments. In “hard” science controlled experiments define their discoveries, in “soft” science computer models and “normative” ideas define their output. The closest relative climate “science” has is either economic or market science. Both economics and market science rely on what are called “econometric” studies that are statistical means to “control” for factors outside of a laboratory environment.  Because social sciences can’t put a society, economy or market in a laboratory, they needed to develop statistical techniques to mimic what the hard sciences were doing in the laboratory.

The closest relative climate “science” has is either economic or market science. Both economics and market science rely on what are called “econometric” studies that are statistical means to “control” for factors outside of a laboratory environment.  Because social sciences can’t put a society, economy or market in a laboratory, they needed to develop statistical techniques to mimic what the hard sciences were doing in the laboratory. Those software packages are as abundant in marketing firms as they are in social science departments of our universities. My understanding is that the first climate models were simply edited financial market models.

While I don’t have experience building climate models, I have a great deal of experience building econometric and financial market models, in fact, it is my background in building these kinds of models that got me interested in this climate “science” debate. I was simply shocked by the obvious flaws in their theory and models. None of the stuff I was reading would have gotten anything but an F- in an econometrics 101 course, and yet climate “science” was using terms like 95% certainty. The only 95% certainty in the field of climate “science” is that their models are failing and will continue to fail miserably.

Unlike the gravity experiment detailed above, climate science can’t run controlled experiments. The global climate is simply too complex and has too many variables. To put it simply, you can’t put a global climate in a test tube. Whereas most hard science is done with single variable models where experiments isolate the variable being studied, all valid climate models are multi-variable models, and those require a separate set of statistical calculations called multivariable linear regressions. To have a valid multivariable model you need to have to meet specific criteria like including all the major/significant contributing factors, one factor can’t be a function of the other (multi-collinearity), the variation of the data needs to be constant (no heteroscedasticity), the data can’t be a function of the previous data (no serial correlation).

The classic example of a multivariable model is weight-loss. Like the climate models, it is a simple input/output model. Any valid weight-loss model would have to include caloric intake (diet) and caloric expenditure (exercise). If the model just included diet, you might get an R-Squared (explanatory power of the model) of about 40, meaning that 40% of the variation in weight can be explained by diet. If you add exercise, the R-Squared may go up to 80. Then you can add in things like sex, age, starting body fat, etc etc. Each added factor will increase the R-Squared (adjusted R-Squared) if it is a valid factor and provides some explanatory power.Screen-Shot-2017-03-10-at-8.23.06-PM

The key to these models is that you have to include all major significant influencing factors. To identify all the significant influencing factors you have to understand them and understand their interaction with the other factors in the models. By the IPCC’s own admission, the vast majority of significant climate factors are categorized as having a “Very Low” level of scientific understanding (see graphic at the top of this article). Claiming a “consensus” and “certainty” upward of 90% is simply inconsistent with having a “Very Low” understanding of the majority of the factors in your model. Climate “scientists” may claim that all they want, but any real scientists knows the claims are pure hogwash.

Even Ex-President Obama admits the “certainty” is up for debate. He recently stated that:

“Ninety-nine percent* of scientists who study climate change carefully . . . will tell you that it is indisputable that the planet is getting warmer and the only real controversy is how much warmer will it get.”

That statement is a non-sequitur when put in the context of CO2. First, just because it has been warming since the end of the last ice-age doesn’t mean CO2 is the cause, and “how much warmer will it get” covers a very very wide range from the meaningless to the catastrophic. For those nonsensical, confusing and meaningless comments, Ex-President Obama got paid over $3 million. Anyway, Ex-President Obama himself refutes the claim of the “consensus” on catastrophic warming.

Please like, share, subscribe and comment.

Advertisements

1 thought on “How Can You Have “Certainty” When So Many Factors are “Unknown?””

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s