12-11-2013, 02:13 PM

Hello All,

I would genuinely like to know, what do you think that it is that the IPCC's climate models actually do?

What do they model, and what are they projecting?

There are no wrong answers to my mind as such, I just want to know what people think the IPCC'S climate models model.

Richard Patton answered on fb that -

They are modeling the Navier-Stokes differential equations for fluid flow. These are some of the wickedest equations to work with. Mostly engineers use empirical tables as general solutions to these equations do not exist.

The GCMs are coupled land and sea computer models based on these equations. They are not "solutions" in the sense of a mathematical equation - they actually run the equations with a bunch of parameters. They make a 3d grid and then run the inputs and outputs of the grid calculation - step at a time with some delta-t (change in time).

Of course, as they assess the changes the grid cells don't all balance out - that is there is always some residual energy that they simply spread out. The last I looked into it the amount of residual energy was on the order of the effect being measured. This was all I needed to see.

Folks who are more experienced than I regarding compute solutions of differential equations argue that the problem being solved is "ill-posed" and thus not a valid procedure. I cannot validate this claim myself but the argument was pretty persuasive.

In either case, any modeling problem is only as good as what it can predict "out-of-sample" - that is it can only be verified to the extent it actually predicts future events accurately (hindcasting is not a valid verification procedure) and even here only provisionally. Obviously these computer model are in a big fail mode at the moment - they are statistically significantly wrong at the moment in their predictions.

I would genuinely like to know, what do you think that it is that the IPCC's climate models actually do?

What do they model, and what are they projecting?

There are no wrong answers to my mind as such, I just want to know what people think the IPCC'S climate models model.

Richard Patton answered on fb that -

They are modeling the Navier-Stokes differential equations for fluid flow. These are some of the wickedest equations to work with. Mostly engineers use empirical tables as general solutions to these equations do not exist.

The GCMs are coupled land and sea computer models based on these equations. They are not "solutions" in the sense of a mathematical equation - they actually run the equations with a bunch of parameters. They make a 3d grid and then run the inputs and outputs of the grid calculation - step at a time with some delta-t (change in time).

Of course, as they assess the changes the grid cells don't all balance out - that is there is always some residual energy that they simply spread out. The last I looked into it the amount of residual energy was on the order of the effect being measured. This was all I needed to see.

Folks who are more experienced than I regarding compute solutions of differential equations argue that the problem being solved is "ill-posed" and thus not a valid procedure. I cannot validate this claim myself but the argument was pretty persuasive.

In either case, any modeling problem is only as good as what it can predict "out-of-sample" - that is it can only be verified to the extent it actually predicts future events accurately (hindcasting is not a valid verification procedure) and even here only provisionally. Obviously these computer model are in a big fail mode at the moment - they are statistically significantly wrong at the moment in their predictions.

The whole aim of practical politics is to keep the populace alarmed

(and hence clamorous to be led to safety)

by menacing it with an endless series of hobgoblins, all of them imaginary.

H. L. Mencken.

The hobgoblins have to be imaginary so that

"they" can offer their solutions, not THE solutions.

(and hence clamorous to be led to safety)

by menacing it with an endless series of hobgoblins, all of them imaginary.

H. L. Mencken.

The hobgoblins have to be imaginary so that

"they" can offer their solutions, not THE solutions.