«
»

I Will Accept This Description

November 29, 2011, 5:41 pm

I normally object to the ways in which global warming alarmists portray skeptics.  But I will accept this from Phil Jones

They [skeptics] mostly look at observation papers and ignore modelling ones, as they believe by default models are wrong!

Models are nothing more than scientific hypotheses programmed into computer code.  As such, I must admit to finding papers that merely model various hypotheses (generally in a very nontransparent and non-replicable sort of way) to be the least interesting of all possible papers.   It is far more interesting to see someone lay out their hypotheses and attempt to justify them with observational data.  In climate, the prevalence of modelling tends actually obscure this discussion, as we don't always see all the relevant hypotheses that form the foundation of a model, and even when we do, we usually don't see the details of its implementation  (which can be as important to the results as, say, the exact wording of a poll question).

So, yes, if Dr. Jones wishes to defines the sides in this debate as skeptics whose science is driven by observational data and alarmists whose science is driven by computer models no one has seen or replicated, I will accept those definitions.

Tags: , ,
Category: Climate
  • me

    Honestly, I blame lack of sound math education - a world in which even scientists truly believe that a model with a huge number of parameters is any good as a tool for prediction is just sad (background for the less scientifically minded: if you give a model sufficient state, it's likely to be able to produce any type of behavior you want and fit any kind of data; it's useless).

  • Larry Sheldon

    He finally says something truthful! I'm sure a reviewer will make him fix it.

  • anoNY

    Jones et al., the first psychohistorians...

  • MJ

    And when your model doesn't fit the observed data, it is your model that is wrong.

  • Ted Rado

    I have stated this many times before: A model that is not based on first principles and contains NO fudge factors is useless and dangerous. The climate models are far from this necessary situation. How anyone can argue that we should destroy our industrial economy based on these models is astonishing.

    As I have also pointed out before, there are no viable large scale alternatives to fossil fuels, so there is nothing we can do (other than move north) even if the models are correct. Meanwhile, our sterling leaders in DC continue to piss away money based on the premise that a) the models are accurate and b) someone will magically come up with an alternative energy source. Historians a few decades in the future will have fun figuring out how we could have been so stupid.

  • Cville

    One of the best modelers I know told me that, "all models are wrong but some are useful". I'm not sure the current climate models have gotten into the later category.

  • IGotBupkis, Sailing the Economic Seas Betwixt Scylla And Charybdis

    As someone with both the computer knowledge and the mathematical knowledge needed to implement (though probably not to devise without more expert assistance) such models, I can easily say with reasonable certainty that existing models are not just wrong, but blatantly wrong.

    "The validity of a science lies in its ability to predict".

    By this alone it is self-evident that AGW as a theory has nearly zero validity. It predicted more and worse storms, more and worse warming, and more and worse climate impact. It has predicted warming atmospheric layers, warming oceans, and consistently less polar ice. It expects no "warming" on other planets as a corollary expectation.

    Time and again it has been utterly and completely repudiated.

    Q.E.D.: The models have near zero reliability.

    P.S., the real, proper test of a model is to plug in 10 or 20 year old climate data into the model, and see how long it takes for the model to diverge from the very real answers that followed. When they can do this and get consistent numbers that match the longer-term results with the real-world results, THEN the "science" of AGW will be established. But if you're going to claim 20+ year accuracy to a tenth of a degree in the future, you need to be able to show you can produce that from 20 year old existing data from 20 years past to NOW.

    -------------------------------------
    This seems remarkably **DUH**.
    ------------------------------------

  • IGotBupkis, Sailing the Economic Seas Betwixt Scylla And Charybdis

    >>> anoNY:

    Jones et al., the first psychohistorians…

    I'd agree with the "psycho" part, anyway.

  • IGotBupkis, Sailing the Economic Seas Betwixt Scylla And Charybdis

    @me: (background for the less scientifically minded: if you give a model sufficient state, it’s likely to be able to produce any type of behavior you want and fit any kind of data; it’s useless)

    This isn't exactly true due to the chaotic nature of the dynamic models in question. There is only a limited number of iterations that a fudge-factor & parameter-tweaked model can mimic a chaotic/dynamic system if it does not actually match the inherent chaos of the dynamic system in question. This is usually referred to as "the butterfly effect".

    But we are far from having any sort of models which are even remotely accurate as it is. They are, in fact, laughably inaccurate. To implement major economic and environmental policy decisions on them is flat out ludicrously stupid.

    Further, the history of issues with complexity and government environmental "solutions" is covered by Michael Crichton in a very well-written and easily understood speech:
    Fear and Complexity and Environmental Management in the 21st Century
    Unfortunately the text version of the speech has become an un-speech thanks to the obvious libtards that have gotten control over his estate... both of the two Anti-AGW speeches (the above and "Aliens cause Global Warming") have "conveniently" disappeared from the available text on his site, and even the archived versions have been hunted down and disappeared.

    I did chase this copy down

    NOTE: These are both long, but VERY worth reading and watching. The text version of the Complexity speech is good because you can start and stop it fairly easily and still follow it.