Soren wrote on Jun 7^th, 2010 at 8:08am:

Now hang on there old chap! A short while ago it the problem was that climate forecasting is wrong because it's all based on these new fangle model things, and modelled predictions were, you insisted,  wrong per se. But now, apparently, modelled predictions can be right, but they are only reliable once the models have been thoroughly tested! Well at least we can agree on that, what! But the problem we have now is that you seem to have retreated, old boy, into one of those comfortable leather armchair areas of subjective assessment: for it doesn't matter how accurate or otherwise climate model predictions may turn out to be, one can always say the model needs more testing, no? Now as it happens, I've personally, always been a bit suspicious of that universal law of gravitation business. Who's this Newton chap think he is anyway? Hasn't earned his stripes in my book. Besides, I dropped another one of those damned cups this morning and the confounded thing flew off into the damned ceiling!

Quote:

Yes:
Quote:

What's the problem? As I demonstrated with gravity, all scientific prediction is based on modelling. That's how it works. By inferring that "all we can do is ..." you are implying that this is somehow less rigorous than "real" science. You're wrong.

Quote:

No. You didn't. Which is just as well because it's wrong. Actually, you were insisting that science only applies to "physical objects", remember?

Quote:

Do you have any evidence that climate model assumptions are not empirically tested? (Here's a hint, the answer will be no.)

Quote:

Ummm, sustained cooling, to state the bleeding obvious. Go be a good boy and find some evidence for it will you. But do be wary of the existing attempts. I would like to think that one of the things you've learned in this forum is that they're all a crock. I wish you good luck, and keep in mind that I and many others will ge genuinely and sincerely elated if you should succeed. As, no doubt, will my children, and their as yet unborn children. But I'll stop. You've got work to do.

Quote:

Well done, you've figured out why climatology relies so heavily on modelling. That doesn't mean the assumptions aren't subject to empirical testing though, does it.

The only conflating I can see is that you seem to be conflating the definitions of conflate and confuse.

Nevertheless, there are very few fields in which I am actually qualified to talk about, but it just so happens that one of them is theoretical modelling. It was you who introduced computers into the discussion. I have been since wondering what exactly it has to do with anything. It's as though you think a 'computer' model is somehow mystically different to any other form of theoretical model.

A computer is a tool you can use to construct a model. Other tools you might use are calculators, pencils and a shitload of butcher's paper. But a theoretical model is a theoretical model, regardless of how it's built. When I predict that upon release my cup will accelerate towards the Earth's centre of mass at approximately 9.81 metres per second squared, I am using a theoretical model just as surely as if a climatologist predicts that the West Antartic ice sheet will be approximately 25% depleted in 2065.

Neither Newton's universal law of gravitation, nor the Antarctic climate in 2065 'depend' on computer modelling. They don't 'depend' on any sort of modelling whatsoever. The predictions that we make with regard to gravity and Antarctic climate do depend entirely on modelling. I'm sure you suspect that if you hold a cup at eye level it will fall when you release it. But you can't prove it in advance of doing it. You cna't actually be 100% certain what is going to happen. But your brain can make a very well defined prediction of what is going to happen. How? With the model it has built to help it predict what happens when you let stuff go.

All of our predictions are built with models. It's the only way we can predict anything because we cannot, by definition, have any kinetic experience of, or direct contact with the future. The phrase 'model based prediction' is a tautology. It is a nonsense to attempt to invalidate or call into question the validity of a prediction purely on the grounds that is is based on a model. And the question of what tools were used to build a model is entirely irrelevant.

If you want to invalidate a prediction, you can test it empirically (which I would seriously discourage in the case of the Western Antarctic ice sheet) or you can test the modelling behind the prediciton. Please feel free to do just this, it's a vital part of the modelling process and it's damned expensive getting someone to do it. So if you want to do it for free go right ahead. No one will mind a bit. No one. But do be a bit more specific. To just say 'Oh it's a computer model', is lame at the very best. More to the point, it demonstrates a complete misunderstanding of not just how science works, but how the human mind forms conclusions.

Exactly! You've got it (Almost: the picture of a pipe is not a prediciton, but I guess you could use it to predict what a pipe might look like).

You could add that a climate forecast is not the climate.

If you have the actual result, you wouldn't need to predict it, would you. But we don't have the actual result, terain, pipe, weather, climate, falling cup, whatever. So we use models.

I won't smoke the picture, but the picture might be useful when I am next looking for a pipe to smoke. And as it turns out, I am looking for a pipe to smoke: I've heard that you are able to predict the winner of last year's Melbourne Cup with no more information at your disposal other than the actual result! Outstanding. After we smoke that picture, umm pipe, we'll collect our winnings and hit the town!

When they were handing the wits out you must have got there a bit late, so they gave you a crappy set.

But at least you beat Grendel, for whom they could only spare a half of one. So you should aim a bit higher than to emulate his ripostes.

Paella wrote on Jun 9^th, 2010 at 8:16pm:


I think Grendel was actually starting to think before he left the forum.  I almost felt regret after explaining that he didn't really understand the basics of radiative forcing and that he was arguing from an untenable position. I'll say this for Grendel - in his last few posts he was actually trying to understand the basics.

I actually don't think Soren is dim-witted by any means. It's just that the Energy companies have succeeded extremely well  in spreading confusion among the general populace.

Paella wrote on Jun 10^th, 2010 at 11:09pm:




Zzzzzzzzzzzzzzzzzzzzz..........................