Science of Global Warming And Possible Man Made Causation

Now I'm not quite sure what is happening here, but I think the accusation is that such figures are set at values by the scientists to give the results they want from the models.

Isn't that one way to state something that we all know is going on, which is that the models are "tuned" until they give a result that resembles what is desired, then the runs that didn't work are thrown out?

"Tuning" is done with the parameters. It would be interesting to know what parameters were "tuned" and which are not.

As an example, does modeling start with a presumption of a certain climate sensitivity, then tune everything else until that presumption is "truethy?"

Surely not...
 
Isn't that one way to state something that we all know is going on,
You don't speak for me. Don't forget it. I consider this statement highly impolite.

which is that the models are "tuned" until they give a result that resembles what is desired, then the runs that didn't work are thrown out?
The source code is right there. Let us know when you've downloaded it and run it. Until then, I see nothing from you but a lot of talk, and many insults.

Software: ftp://ftp.giss.nasa.gov/pub/modelE/modelE1.tar.gz
Instructions: http://www.giss.nasa.gov/tools/modelE/
Let us know when you've got some simulation results to share.

Until then, why don't you just buzz off? Basically all you do is lie and insult people, and it's boring.
 
You don't speak for me. Don't forget it. I consider this statement highly impolite.

The source code is right there. Let us know when you've downloaded it and run it. Until then, I see nothing from you but a lot of talk, and many insults.

Software: ftp://ftp.giss.nasa.gov/pub/modelE/modelE1.tar.gz
Instructions: http://www.giss.nasa.gov/tools/modelE/
Let us know when you've got some simulation results to share.

Until then, why don't you just buzz off? Basically all you do is lie and insult people, and it's boring.

Thanks, but I'll pass on buzzing off, and also on running the Fortran.

I've written and ran lots of Fortran code, but don't see a clear relation or possibly any relation between running code for one particular model, and the specific questions here.

But it is indeed quite handy to be able to just say the Model says so, so it must be right, etc.
 
So basically, you don't know how, and you don't want to admit it.

Thanks. We'll keep your level of knowledge (zero) in mind.

I'll take that as I had it right with my initial comments on the derivation of the alpha, thanks. It seems like a bit of a touchy subject, alpha, once one digs into it a bit....

For anyone who wants to actually run a climate model at their PC without installing and becoming knowledgable of Fortran, just download this.

Very nice little NASA package, about 70mb, allows running tests of doubling CO2 and the like - intended for educational applications.

No, I haven't looked at it to see where they get their values for alpha.:D
 
I still see zero.

Please point out alpha in the source code, the link to which is above.


first "it's in the IPCC somewhere"
Then "oh no, guess its not"
Now "here's a million lines of code, it's is there somewhere!":)
 
Thanks for confirming my guess that you have absolutely no idea what's going on.

Alpha doesn't appear in the source code.

You're not very good at this, are you?

Why don't you tell us all WHY alpha doesn't appear in the source code for the climate model. For extra credit, indicate what alpha is for and where it came from.
 
Syntax aside, empirical is the very opposite of modelled. It means direct observation.

Thank you. I was unaware that I had written anything contrary to that.

The "accusation" is apparently mhaze's that alpha is a made-up number. Where model predictions come into it is anybody's guess (given the syntax problem).

The accusation was not mhaze's.
 
OK, let's try this again: alpha in that section of the IPCC report is not usable in any serious climate model because it is a composite representation of the overall effect of CO2 throughout the atmosphere, given to allow consumers of the report (not the scientists doing climate research) to make a guess at how much change in radiative forcing there will be for a given change in CO2. That this number would change over time, both as our knowledge of the characteristics of the atmosphere and as the composition of the atmosphere change, is unsurprising.

Models don't use a composite representation, they use an exact representation, for a layer of the atmosphere; the layer size (or more properly, looking at the code, the number of layers) is a parameter of the model. Because the optical depth changes with temperature and pressure, and because temperature and pressure change as one moves upward, the value of the optical depth will change, and it is these values that get used in the models. Obviously, the more layers are used, the more accurate the model will be.

It is therefore unsurprising that the number given in the context of the IPCC report would be derived from the models, and not be present in any of them. It is a rule of thumb, not the actual basis of calculation. The actual basis of calculation is the HITRAN database of spectroscopic data for gases.

It was my mistake as much as Slime's thinking that number was part of the models. I should have done better research and cleared that point up properly before now.
 
Now I'm not quite sure what is happening here, but I think the accusation is that such figures are set at values by the scientists to give the results they want from the models.

That's basically what I saw during my search for 5.35. Minor "adjustments" are allowed in modeling to bring predictions to empirical observations but they have to be accross the board. IOW, CD could not be "corrected" any more than any other GH gas. Still, I don't know what's going on with that constant.

I might ask him again, since people seem to be genuinely interested in how things actually work, rather than just beating up on him for being a commie who wants to kill their babies and steal their money.

You say that as if it's a bad thing! If true, this guy and I could really get along. ;)
 
Assuming that I am on the same page as the rest of you (and I didn't see this linked before)

The original paper with the radiative forcing constant for CO2 with a value (alpha = 5.35) can be found here.

Myhre et al. 1998. New estimates of radiative forcing due to well mixed greenhouse gases. Geophysical Research Letters Vol. 25, 14, pp. 2715-2718.

Isn't this paper the primary source for the value?
 
I'm not the one running around calling everyone liars, though.

You're smearing all climate scientists by accusing some of manipulating their results (for unspecified reasons), without naming names. You could stop that by, you know, naming names and telling us what they're accused of.
 
Good job, Pipirr. Note the first paragraph of section 2 where the HITRAN database's use in the radiative model is discussed, and the fact that the coefficient is clearly shown to be derived from the model, and is not part of it. That should close the discussion on this matter.
 
I'm SURE we're going to get into how radiative codes work, so here's the deal:

First, you've got spectroscopic characteristics measured at various temperatures and pressures and concentrations and so forth. Then, you've got to figure equations to describe how those characteristics vary in between the measured or first-principle calculated values. Then you have to make a model of how the parameters will change through the atmosphere, as the pressure and temperature and concentration change with altitude, so you're going to be using PDEs to describe that. Then you have to make a model of how the atmosphere looks from that, probably by defining homogenous layers rather than attempting to make a continuous function vary over another continuous function, so you can use numerical simulation to do it. Then you have to account for the fact that the Sun doesn't just sit up there in the middle of the sky, it moves, and as it does, various refractive indices change, and the path length through the various layers changes too. And finally, you have to put some feedback paths in there, so that the larger climate model can feed whatever variations in temperature and pressure and concentration back in, to get more accurate radiative calculations.

That must make for a pretty full morning.

So when I see someone say something like, "Hey, Hansen is just lying about this," I kind of get the idea that they're not quite clear on the fact that there's like fifty or a hundred people all doing all this different stuff to make the model, and they know so much that there's no way Hansen could like, sneak something in there or anything, so they'd ALL have to be lying together and never make a mistake or someone would find out. Not only that, but there's a bunch of other groups out there doing the same thing, and if the results are way different, you're going to have a big controversy and whoever's right is gonna get the gravy and whoever's wrong is gonna get hosed. That's how science works.

Well said.
 
Isn't that one way to state something that we all know is going on, which is that the models are "tuned" until they give a result that resembles what is desired, then the runs that didn't work are thrown out?

Who do you mean by "we"? The Illuminati? And how do you all know it? Wouldn't this sort of thing be kept secret?

Assume, for a moment, that this was going on. Let's say, for instance, the Hansen et al 1988 model was tuned to produce the climate forecast it did (leaving aside why it was the desired prediction). The likelihood would have been that the model would not have performed well in real terms - unless it was possible to tune it to what was actually going to happen, which implies an existing ability to predict climate and so can be dismissed.

In fact the model has performed very well over twenty years. Unless you want to put that down to luck and wasted effort it doesn't seem that tuning for a desired outcome was involved.

It would have been impossible at the time anyway, there simply wasn't the computer capacity to run the models dozens or more times just to tune them. 1988 was Bronze Age, where 1978 (punched-cards and FORTRAN) represents the Neolithic. It just wouldn't have been logistically feasible.

Generalising from this, where are the tuned model predictions that you all know have been produced that have gone wildly wrong? If they're so common you can surely point to one or two. Unless you think coincidence has really been stretched, with all those tuned models lucking-out.
 
The actual basis of calculation is the HITRAN database of spectroscopic data for gases.

Observed and confirmed empirical data, in no way susceptible to massaging. Just to add some stress to that point.

It was my mistake as much as Slime's thinking that number was part of the models. I should have done better research and cleared that point up properly before now.

Thanks for doing so, and for sharing it so effectively. I'm learning a lot.
 
Assuming that I am on the same page as the rest of you (and I didn't see this linked before)

The original paper with the radiative forcing constant for CO2 with a value (alpha = 5.35) can be found here.

Myhre et al. 1998. New estimates of radiative forcing due to well mixed greenhouse gases. Geophysical Research Letters Vol. 25, 14, pp. 2715-2718.

Isn't this paper the primary source for the value?

It does answer the question:

Where does the 5.35 come from?

Slimething's link noted in post #617 is IPCC 2001 does use the Myhre 1998 number in the first line of table 6.2.
 
There's a pay download button, right next to it is a place where another button appears when it's done counting down. You have to wait for the countdown to finish. It's kind of irritating, as is the popup ad they make you close, but not too bad. Just be patient until the free download button comes up.
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom