And this is how climate change is 95% certain

Big headline over at WtD: We’re 95% certain. Did you ever ask yourself how they came to this conclusion? What empirical testing in the real world was done to verify this 95%? Or was it all based on models proving models (circular reasoning–you cannot prove a model with a model). More important, is 95% that huge a confidence level? Are two standard deviations from the mean (assuming we have a mean, it’s calculated correctly, etc) significant enough for something as important as climate change?

I found a blog article that answers some these questions very well:

95 percent confidence: in HEP vs IPCC
When I saw some reports about the IPCC’s 95 percent “certainty” that the global warming is mostly man-made, I couldn’t avoid thinking about the huge difference between hard sciences (such as particle physics) and soft sciences (such as the contemporary climatology).

Continue reading here:
http://motls.blogspot.com/2013/08/95-percent-confidence-in-hep-vs-ipcc.html

For those of you who don’t do statistics, etc, thus far the hotspot in the troposphere has not been found (the statement there was “it’s not essential to the theory) and warming flattened off for over 15 years (now it moved to the ocean–or if you’re Kevin Trenbreth, he seems to have just thrown in the towel with: it will get hot some places and cold others). Climate change has been replaced with “extreme weather”–a term that is anything the media decides it will be, allowing any event to “prove” the predictions. If any of you have ever watched a “psychic” do their thing, when one prediction fails, the claim is the vision wasn’t quite clear but now they can say with confidence that “X is true” and will happen. They usually move the time frame forward (you know, like now the warming will be by the end of this century, at which time most of those predicting will be dead) and hope you will forget the original prediction. Climate change “science” has no better a record than a standard fortune teller, yet proudly proclaims 95% certainty.

The only thing we can predict with certainty is if we continue to allow people to think cheap predictions are somehow science, we will head back into the dark ages of superstition and psychics with nothing to stand in the way. Science will be buried. Maybe that’s preferable for some–it requires no thought, plus it’s faith-based so no one can counter it. Why let a little thing like reality ruined an otherwise perfect system?

If someone would like to explain in very clear statistical language how the IPCC came to the 95%, please, feel free to share. It would be helpful if you could show that actual data used and the statistical techniques, fully explaining the models, etc.

Scientific badger

Scientific badger

Advertisements

2 comments on “And this is how climate change is 95% certain

  1. Actually, not much hindcasting is done so far as I can tell. I will look for an example of a model proving a model.

    I am not being ironic, or I don’t think so. I would go for 3 or 4 standard deviations out. 99% and I need to see how that number was figured. Remember, this is predicting the future.

    Medical trials and houses burning down are not based on modeling with a computer, though medical trials get close, nor is standing in the middle of the road in front of a car based on computer models. I remember as a child being part of the oral polio vaccine test–one that would NEVER be done now, because the assumption would be that not giving it to children could be depriving them of a valid vaccine that was easier to administer and was more apt to be taken because of that. However, had the vaccine been useless, thousands of children would have been susceptible to polio–i.e. unvaccinated.

    Interesting, you would advocate using a medical treatment that had positive results in fewer cases than random chance. You keep telling me you understand the scientific method, but you advocate giving medicines that are no better, or even worse, than the placebo used. Would you agree with Steve Jobs’ choice to use herbal treatment early on for the prostate cancer that eventually killed him? Apparently, he believed in it and sometimes it may work (an actor once made that claim and he is still alive–so maybe?). Where then where do we draw line between the Magic Show Exlisor and the scientific one?

    As for insurance, I don’t care if someone has it or not. That’s their choice. If their house burns down, they can always beg on the internet. Getting money after the fact to rebuild is cheaper for them and quite effective. Personally, I have insurance. That’s my choice, now that I own my home. If the person has a mortgage, they don’t usually have any choice. Mortgage companies want that 99.9999% certainty they won’t lose their asset.

  2. youkipper says:

    “Or was it all based on models proving models”

    Your tone suggests this is a common method for proving models. Have you any examples because I can think of none and I’m pretty sure most climate models use historical data and hind casting for testing?

    “More important, is 95% that huge a confidence level? Are two standard deviations from the mean (assuming we have a mean, it’s calculated correctly, etc) significant enough for something as important as climate change? ”

    I am less than a little surprised by this statement. ‘Very’ might be more appropriate but I can’t tell if you are really serious or just being ironic.

    If you really are serious, what level of confidence do you think is significant enough?

    Given that the chances of someone’s house burning down within a few years is likely to be many times less certain than 95%, I doubt that you would be advocating they not have insurance, or given that medical trails often show less than half of patients improving as a significant result for treatments of illnesses, I can’t believe you wouldn’t offer such treatment so I really would like your reasoning for questioning a result that most would consider almost certain. I certainly wouldn’t be standing in a road if that were my chances of being run over.

Agree? Disagree? Leave a comment!

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