What's the point in climate models?

Question

It strikes me that climate models are a pointless waste of time, no matter what your viewpoint is, because, as the climate is so unbelievably complex, it is impossible to make accurate predictions of the future.

This is shown in the huge variation in the IPCC models: more than 4oC variation - they can't all be right and no-one knows which one is, so the number any one comes out with is all but useless.

You might say any ONE is useless but TOGETHER, we can build a picture of the future, but that's saying "the central value for the warming without error bars, is 3 degrees give or take a couple of degrees": it's just pointless.

Also, when models have predicted way over the mark of real temperature change, we just sya, "oh well we didn't know as much about modeling 10 years ago as we do today", but what's the point in a prediction you never assess the accuracy of, because in the future, when we, inevitably, miss the mark we will once again say we didn't know as much about it

Answer 1

Dana said:

"So what exactly makes you qualified to conclude that the climlate is too complex to model accurately? How many PhDs do you have?"

You don't have to have a PhD to know that science knows very little about climate--I'm pretty sure a third grader could tell you that. Thats not to say we don't know anything, on the contrary, we have advanced leaps and bounds with regards to knowledge of our climate, but there is still so much we don't know, or know very little about. Now if there are very large gaps in our knowledge of what drives and influences climate, then how could we expect to model climate accurately? We can't. We can only hope to create a model that could get fairly close to observations at some point in time (even if it is mostly by chance), if some unknown plays out in a certain way (scenarios).

And contrary to popular belief, hind-casting is not a test for accuracy--models are tweaked to fit observed data (and this is not devious in any way, as it is common practice in the modeling world).

And regarding the realclimate post: observations played out between scenarios B and C, though somewhat closer to B. This shouldn't be held up as proof of accuracy by anyone--C is the scenario where CO2 emissions stop in the year 2000, and both B and C include several volcanic eruptions that lowered the temperatures and slowed warming, and we have only had one (Pinatubo). Doesn't sound too good...

Answer 2

Most of the difference is due to different scenarios regarding CO2 emissions in the future. More CO2, higher warming, there's nothing strange about that. I seriously don't think the world leaders will be dumb enough to allow emissions to rise as much that it will lead to 4-6 degrees increase or more this century, but IF they do, it would indeed be catastrophic for most life on earth. The message here is that it's very much up to us how much temperatures will rise in the future. (But of course not only up to us, since we're unable to control all variables that affect climate.)

Actually climate models have been able to predict future climate as well as reconstruct past climate and that's how we can say they are accurate:

http://www.logicalscience.com/skeptic_ar...

Answer 3

These models are extremely useful if you want to propagate a lie- just weight the various factors to produce a desired outcome and you too can deceive the public!  Nobody seems to have noticed that none of these models have been able to reproduce past events...

Answer 4

So what exactly makes you qualified to conclude that the climlate is too complex to model accurately?  How many PhDs do you have?

How do you explain the fact that climate models have been extremely accurate so far?  Is it just blind luck?

http://www.realclimate.org/index.php/arc...

Also, the range in climate predictions is because we don't know what will happen with future CO2 emissions.  The lower warming predictions are under a scenario where we drastically reduce our emissions.  The higher warming scenarios are if emissions continue to rise.

It's not that the models vary greatly in their predictions, it's that they have a range of different scenarios.

Answer 5

They enable us to make an estimate of the problems we face.  As time goes on those estimates are refined.

The IPCC's error estimates are way conservative, to accommodate differing views within the group.  This graph shows pretty well what the real uncertainty is, and it's not several degrees.

http://www.globalwarmingart.com/wiki/Ima...

Answer 6

climate models are a good thing.

some work better than others.

as computers get faster, and have more capacity, models can become more precise.

testing models is quite possible.

you enter the data that was available from 30 -10 years ago, and let the model predict what's happened in the last 10 years.

the closer it comes to reality, the better we can assume the model is.

now the fact is, you don't want to hear this.

you'll go and write again about how useless models are.

the most i can hope for is that somewhere within you, there is something that will realize you're wrong.

maybe someday, the pretty obvious truth will become sufficiently obvious to you that you'll change your mind.

the only question is, how far off that day might be.

Answer 7

Without computer models how are they going to scare the public of catastrophic temperature rises?