Quote:
Originally Posted by Overdog
Perhaps you misunderstand the nature of my general critisism of of climate models. Let me post the following, which I agree with, to see if we can determine where our points of view differ.
|
I can see how the tone of my previous posts may have led you to this conclusion, but I do not think I have misunderstood you. However, to be sure, I will summarize below my understanding of your points so we can see if I am accurate in my understanding.
- Climate systems are enormously complex, and have hundreds of variables to consider
- Models use different programs to achieve their results (the maths are not consistent from model to model)
- Models are often adjusted manually, hence introducing another source of potential error
- The across-model outputs are not always aligned
- Climate is a chaotic system
Is this a fair assessment? If not, please let me know if I've misrepresented or misunderstood you and we can correct that relatively simply.
Quote:
Originally Posted by Overdog
Now consider the nature of chaotic systems. Can we agree that Climate is a chaotic system?
|
Well, almost. Climate and weather are, in fact, the two most common examples of chaotic systems since they evolve with time and are highly dependent on initial conditions. I think we both agree that it's fair to suggest that these models have many parallels and characteristics of a mathematically defined dynamic/chaotic system.
The bigger point, in my view anyway, is the distinction between modelling weather versus modelling climate.
As my analogy about the waves on the beach in a former post alluded to, it's much simpler to predict trends accurately over long expanses of time than specific and individual moments or weather events. I may not be able to tell you exactly where on the wall the next incoming wave will hit, but I can tell you the average on that wall that the waves will hit throughout the coming year with enormous validity.
This is how we model climate. We are talking about global yearly averages, not what the weather will be like on Thursday, November 27 at 3:42PM Central time in Waupeton, Iowa.
This is why a little bit of error in those initial conditions is less worrisome. Is there noise? Yes, absolutely. I concede that, and I'd either be a liar or really stupid if I tried to suggest that models are perfect (and, I don't classify myself as either of those things). However, that noise is reduced rather significantly by modelling the data over years instead of exact moments and locations, especially since the accuracy of our inputs gets better all of the time.
Quote:
Originally Posted by Overdog
Now, I ask, how well do we know the initial conditions for Climate? Do we even know all of them?
|
No, but we also don't have to. First, I want you to remember that the data we have is being improved and of increasing accuracy every single day (partially due to the vast denialist web presence, and that I admit is one of the benefits of these groups). Second, we DO know the conditions which have the greatest impact, and we DO know how to accurately project the change in the inputs.
Quote:
Originally Posted by Overdog
I also note that new factors seem to be popping up regularly as we learn more, and that some factors (such as the effects of cloud cover) are simply not well understood.
|
New factors popping up is EXACTLY what we want, as they only make the models better. This is NOT a bad thing. Also, there are scores of people working on those subjects which you have classified "not well understood," so we'll have that information available very soon.
But here's the rub. The more we learn, and the more we add new information to our models, and the better we understand all of the various factors and variables, the more our predictions are confirmed!
That's the key point in my book.
The more we learn and the better we get, the more we confirm the past predictions. The past predictions have been amazingly accurate (especially considering the mathematical issues with modelling chaotic/complex systems to which you've been alluding). Our predictions are only getting better, and we have without fail confirmed with all of this new information that the future is getting warmer.
Quote:
Originally Posted by Overdog
Now all of this is not to say that efforts to model climate are not useful, but until those models can demonstrate Grassl's four capabilities, like Grassl, I see little justification for considering future climate predictions based opon them as particulary credible. Their accuracy at modeling past and present climate alone is not sufficient.
|
I get the sense then that no model will ever be good enough for you. You are approaching this with your mind made up that climate is too complex to model. I have shown how accurately they represent past climate as a way to show their validity, and you reject that. Well, they have also accurately predicted the degree of warming over the past decade. Way back when the models were still "in their infancy," they STILL got it really damn close.
I accept your challenges to the difficulty of modelling complex systems, but this is why you need to point out specific models.
It's not that the process of modelling is difficult that should cause us to reject them. It's the margin of error, and we cannot calculate margin of error until we look at something specific.
Quote:
Originally Posted by Overdog
Please do not interpret this post as an argument against Global Warming.
|
I never did, but you must concede that we hear this attack on the models much more frequently from those people who do deny climate change. So, if I've suggested that I have lumped you into that ideological bucket, then this was inadvertent and I apologize.
Back to the models though, NOAA has a great site discussing some of the same points I've tried to summarize above (which includes links to other information).
NOAA Overview - Modeling Climate
And here is a summary for a wider audience done in 2007 as part of a "responding to the denialists" toolkit:
Climate is too complex for accurate predictions - earth - 25 October 2007 - New Scientist Environment
And below is a whole swath of information specific to models and the challenges to them:
RealClimate - Articles on Climate Modelling
(the one below is probably the most recent one which comes closest to addressing your specific concerns, but there are many others):
RealClimate - Butterflies, tornadoes and climate modelling
In sum, I understand the points you are making, but I disagree with the conclusions you have drawn from them. You seem convinced that since climate is a complex system it's not possible for us to make models which work. I say, we'll never be perfect, but we're really close, and we get closer each day.
I hope I've done a fair job addressing your points. I still think we'd all be better served to take this from the abstract level of models in general to a more specific work though, since when you distill it down to the barest essence it's the specific work you are ultimately challenging with your posts in this thread.
) who might be interested:
