The IPCC and The Null Hypothesis

With great fanfare, the UN’s International Panel on Climate Change (IPCC) will release the first part of its 5th full assessment of climate change next week.

Its last major assessment, called AR4, was released in 2007. The new report is called AR5. It took 5+ years and a cast of thousands to create.

The IPCC has ignored a fundamental concept of science in every one of its reports:
The Null Hypothesis“.

Ignoring the null hypothesis is a fatal flaw in all IPCC reports and why they must be rejected outright as science-based proof that humans are the cause of global warming.

Perhaps the reason the IPCC commits such an elementary error in basic science is that it isn’t a scientific organization. It is a political arm of the UN staffed by government appointees.

What is ‘The Null Hypothesis’?

In all things of a statistical nature, such as measuring climate change over time, a null hypothesis is defined and tested.

The null hypothesis must be assumed to be true unless statistical analysis proves it is outside the range of random probability.

That range has a strict mathematical definition. It is two or more standard deviations away from the prediction of the null hypothesis.

In laymen’s terms it means there is a 95% or greater chance that the null hypothesis is wrong.

In that case, the null hypothesis is rejected.

The purpose of the null hypothesis isn’t to prove anything. Its purpose is to eliminate possibilities. In the case of climate change it can be used to eliminate natural variability as a cause of climate change.

An IPCC Null Hypothesis

Currently observed changes in global climate indices and the physical environment, as well as current changes in animal and plant characteristics, are the result of natural variability
–  IPCC null hypothesis example

The IPCC has never ever defined a null hypothesis to build its case for human-caused global warming, even though all climate change is statistically based. That is flawed science.

Above is a reasonable example of a climate change null hypothesis. It is taken from a new alternative report to the IPCC called, “Climate Change Reconsidered II“.

The null hypothesis above for climate change assumes that observed changes are caused by natural variability. If statistics prove it must be rejected, then the theory of human-caused global warming is strengthened.

To go about testing the null hypothesis, compare any climate variable, like temperature rise –  that overlaps times of natural variability with times thought human caused. IPCC AR5 says 1950 is the starting point where non-natural climate change effects begin to appear. Apply statistical analysis to data before and after 1950 and if the result is 95% or greater that the change is 0utside what natural variability allows then the null hypothesis – results are consistent with natural variability – must be rejected.

That, of course, does not prove that human’s are the cause of climate change. It proves that climate change is outside the range of natural variability.

In the world of science, though, that is considered very strong support for the theory of human-caused global climate change, though not proven absolutely.

Since AGW theory attributes rising atmospheric CO2 levels to climate change, then a direct link beyond what cannot be explained by natural variability must be established.

That is a standard the IPCC has yet to meet.

The Error of IPCC Ways

Instead of using the standard scientific approach to support the theory of human-cause climate change, the IPCC takes the opposite and entirely false approach.

From its very first report the IPCC has always assumed all climate change is human-caused and seeks out data and builds mathematical modeling to support it. It’s a tainted approach that assumes true the very thing you are out to prove. That is both unscientific and illogical.

It is exactly the opposite of the scientific method as applied in statistics-based studies. Everything in climate science is statistics based. Therefore, the null hypothesis applies.

The IPCC has never applied the null hypothesis in any of its assessments.

Conclusions

The IPCC has always taken a fundamentally flawed approach with its reports. It assumes true the very thing it claims to be assessing. That doesn’t make sense.

No doubt, that is why the IPCC has always emphasized the 98% consensus, as if opinion defines scientific fact.

A couple weeks ago the IPCC leaked its own report to media-friendly sources. None was leaked to critics.

The only headlines coming out of that is the IPCC is 5% more certain now that climate change is human-caused than it was in 2007.

Now, the IPCC’s “confidence” level is 95% instead of only 90%.

That is the type of reasoning the IPCC has resorted to as more and more contrary science comes out that disputes its claims.

Advertisements

About azleader

Learning to see life more clearly... one image at a time!

Posted on Sep 20, 2013, in Business, Climate, economics, environment, Government, IPCC, nature, news, Politics, science. Bookmark the permalink. 16 Comments.

  1. Mix science with politics and you can produce a weapon of mass destruction.

  2. Will J. Browne

    JOE: Boss, we have a major problem. The storm has caused a landslide. A train full of 500 people is heading straight for it! We were going to warn the driver, but Bob stopped us.
    BOB: I had good reason. I have a null hypothesis. Our trains have being running safely for years, and there must have been landslides in that period. On that basis, my hypothesis proposes that landslides don’t cause train crashes and the train will arrive at its destination safely.
    BOSS: Joe, are you certain the train is going to crash if it keeps going?
    JOE: We have been liaising with the police force, the army, the rescue services and the national transport authority to assess the danger. Due to the storm conditions, we can’t be absolutely certain that the landslide has damaged the line enough to cause a crash, but, given its size and location, there is a 95% probability that it has.
    BOSS: 95%, not 100%. That means we have a one- in-twenty chance it won’t happen.
    JOE: That’s an optimistic estimate. The more we look at it the worse the odds get. There is at most a one-in-twenty chance that it won’t happen.
    BOSS: Can you disprove his null hypothesis?
    JOE: Well, I don’t think null hypotheses get disproved. They are either accepted or rejected. I think we have enough proof to reject it.
    BOSS: But you can’t categorically reject it?
    JOE: That’s difficult to do. At 95% certainty there’s debatably enough evidence to consider it rejected. Higher degrees of certainty are normally only achieved in situations that can be tested in laboratory conditions where you have controls. Anyway, time is running out.
    BOSS: If you can’t categorically reject the null hypothesis then maybe we shouldn’t do anything about it.
    JOE: But there is an overwhelming probability that people are going to die! If we at least get the train driver to slow down it might give us enough time to get the obstacle cleared before the train hits it!
    BOB: That would mean we would have to change our schedules. That’s a lot of work. Anyway, even if the train crashes you can’t be absolutely certain people are going to die. There might even be benefits. We’ll get the insurance for one thing and it is an old train. We’ve run it into the ground and it was due to be scrapped soon anyway.
    BOSS: It does sounds like a lot of work, and I’m due on the golf course this afternoon. Anyway! Now I’ve got a null hypothesis that covers my ass. I can refer to that when big government and the mainstream media start kicking up a fuss. You see, Joe, by agreeing with the experts you made the mistake of engaging in groupthink. Bob, you weren’t afraid to ignore the evidence and make the right decision. Well done, you’ve pulled it out of the bag and saved my round of golf!
    JOE: BUT MY FAMILY’S ON THAT TRAIN!

  3. “In that case, the null hypothesis is rejected. If not, then the null hypothesis must be accepted.”

    That is not correct. If the null hypothesis is not rejected, it is not rejected. It does not follow that you must accept it.

  4. Oh look, if you get as far as the introduction in an IPCC report on the IPCC site you get them considering the null hypothesis

    ———
    2.1 Introduction

    Observed climate change and variability (for definitions, see the IPCC Glossary, Appendix I) are considered in this chapter by addressing seven commonly asked questions related to the detection of climate change and sensitivity of the climate to anthropogenic activity.
    [….]

    Sometimes we provide quantitative estimates of uncertainty, as far as possible the value of twice the standard error, or we estimate statistical significance at the 0.05 (5%) level. This is the appropriate terminology and implies that what we see is very unusual, given the null hypothesis”
    http://www.ipcc.ch/ipccreports/tar/wg1/051.htm

    and here:
    http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch9s9-a-2.html

    and here:
    https://www.ipcc.ch/report/ar5/wg1/docs/review/WG1AR5_FOD_Ch02_All_Final.pdf

    and…..

    • The IPCC says a lot of things. If it actually followed the null hypothesis it would have to reject its own AGW theory given that most of its climate model temperature forecasts fall outside the 95% (more than two standard deviations) range.

      • When you say the IPCC ignored the null hypothesis in every report and “never ever defined a null hypothesis” this seems to conflict with them defining it reports

        here:
        http://www.ipcc.ch/ipccreports/tar/wg1/088.htm

        or charting values for the null hypothesis in graphs here:
        http://www.ipcc.ch/publications_and_data/ar4/wg2/en/ch1s1-4-2-3.html

        To say they would automatically reject their core theory if they did the this correctly is a bit different a statement that implied in your article where you say they didn’t do it at all -ever.

        But I would like to know your educational or research credentials since you claim to see an obvious fatal flaw in a “simple fundamental concept” in understanding how one writes a scientific paper that escaped thousands of Phds that worked on the report in a multi-level review.

    • OK… you are right… the IPCC does mention the null hypothesis. They just don’t know how it works, nor how to apply it. The whole premise of AGW theory is that they say they considered everything else and that they cannot explain the observations and therefore AGW theory must be true. That’s the opposite of the null hypothesis.

      The fact remains that the null hypothesis is already close to disproving AGW theory. In another 5 years or so, AGW theory will have to be rejected because of it.

      • The null hypothesis is simply the assumption that there is no relationship between 2 measured phenomena – its the assumption of the reverse of whatever they are trying to test for. For example “man isn’t causing an increase in greenhouse gases” or “C02 does not cause alter the global climate”. It’s just one check on being true, it does not mean you need to test for and rule out all other imagined possibilities first- for example you don’t need to “consider everything else” in every assertion and rule out that warming is all caused by sunspots or volcanoes or leprechauns etc. etc. (there’s no papers asserting the later either way)

        https://en.wikipedia.org/wiki/Null_hypothesis

        I’m not sure how you can say in your article they misapplied/don’t understand it when you never read any of their applications of the null hypothesis (as you said you didn’t realize they tried) and you’ve also redefined the accepted definition of the null hypothesis. (And the thousands of papers cited in the IPCC report all have their own individual null hypotheses – did they all mess up in trying to use it – without you having to read them- based on your non-standard definition?)

        Even if and when AGW theory wrong doesn’t mean this is where they all messed up – there’s hundreds of other reasons why it could eventually be totally disproved without invalidating any of their individual applications of the null theory.

        Further, saying we should accept your “fact” that man-made climate change (AGW) is going to proven to be false because of your prediction of the future is a bold claim given that every scientific organization in the world has accepted this basic premise as true based on current evidence.

        “Warming of the climate system is unequivocal… Most of the global warming since the mid-20th century is very likely due to human activities… no scientific body of national or international standing maintains a formal opinion dissenting…”
        https://en.wikipedia.org/wiki/Scientific_opinion_on_climate_change

    • Thanks for your perspective. Differing viewpoints always enlighten when they contribute to a greater understanding of a topic.

  5. It is my understanding that we know there has been a high degree of climate variance in the past (i.e., it’s been considerably warmer and colder than it is now).

    I think we all agree that humanity’s contribution to greenhouse gasses is a very recent phenomenon.

    So while we can debate the extent to which the null hypothesis has been applied or misapplied, it just seems that in light of the natural variability we know existed, and the relatively small changes we’ve seen in this eye blink’s worth of time we’ve chosen to focus on – the conclusion that it “must be us” is unsupportable, and with all due respect absurd.

    To suggest we have a sound understanding of something as complex as the climate system at this stage of the field’s development ought to be considered as dubious as if someone said they could predict the stock market with 95% certainty based on a single factor.

    In the case of the latter, our bullshitometers would go off immediately. In the case of the former, we just nod and agree to agree because, you know… science.

    But if you look back at any century at what was considered scientific knowledge, you’ll find much of it wrong and some of it laughable. And yet people think this could NEVER be the case today with climatology… with our tiny sample size, primitive measuring tools, huge gaps of knowledge filled in by guess work, massive and uncontrolled variables and results falling outside 95% of predictive models.

    This is narcissistic self-loathing.a la The Population Bomb. Nothing more.

  1. Pingback: While Congress Is Fighting Over ObamaCare, Obama is Planning His Next Attack On America | Asylum Watch

  2. Pingback: Statistisch zand in de ogen strooien -

Comments and questions are welcomed!

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

%d bloggers like this: