Thanks for the reference. I should always check there first. Surprising that Arrhenius got the same estimate from a back of the envelope calculation 100 years ago, than the IPCC with billions of $ of R&D !

I am obviously new to the field, so please excuse my naivete. I have a PhD in physics, and I’ve done a lot of science myself (56 peer reviewed papers) so I guess I can get a grip on some fairly technical issues (e.g I understand what a logarithm is…). I am a sceptic only in the sense that I think challenging the “mainstream” theory is the best way to make good science (which is why I prefer this blog to RealClimate). If you only “go with the flow”, you just look in one direction, and end up seeing just what you expected to see. My best science was when I was attentive enough to find those little anomalies in experimental results that, when you dig a little, end up revealing something much more interesting. There’s nothing more boring than performing an experiment expecting a result, and getting just that result. Yet, 90% of what’s published is exactly that.

I think with climate research, if you are looking for warming, you will always find warming. If you want to relate it to GHG, you will always find a way. If you plug a number for CO2 sensitivity in a GCM model that you got from past data that ASSUME CO2 sensitivity, you will get the warming you want, but it’s circular reasoning. If the past trend was due to another unknown reason, you’ll never know about it because you’re not looking for it.

Francois

ever since Arrhenius(1896) doubling of CO2 has been a value for climate sensitivity, as the raltionship lis logarithmic, you can calculate the inbetween values using Myhres equation.
http://en.wikipedia.org/wiki/Svante_Arrhenius#Greenhouse_effect_as_cause_for_ice_ages

Er ah Francois, we’re on the same page here. I hope you don’t think I’m promoting GCM’s as the way to go. I’m saying that I think a simple algorithm might be more powerful (In response to your earlier discussion where you created one that was).

The “doubling” of C02 is certainly not my idea, nor do I think it will happen in the short (Relatively speaking here to mean by 2100 AD not PM). It was put forth by the IPCC and is in contention by the skeptics. I bring it up because if we can get a truly predictive tool that can predict past data, we can use it to show better what would happen with a doubling of CO2, which the warmers propose will happen by the end of the century.

OT You’d be surprised, I’m sure there are some still there.

Certainly a point can be made that GCM’s can be useful to learn about regional characteristics of climate change, e.g. will temperature change more in the Arctic than in the tropics and by how much. But on the other hand, if you want to model the mean earth temperature, building a complex model that gives you detailed temperatures everywhere, just to average them out afterwards, may not be the most efficient way. A properly parametrized “energy balance” model could give you results that are as accurate or more. The trick is to identify which are the main mechanisms, and model their behavior. Solar activity is an obvious one, because we get most of our heat from the sun. Can we identify a proper “amplification” mechanism for it? The greenhouse effect should also be taken into account. Details such as the effect of cloud cover can be included in cross-products. That’s the main difficulty of such models: if the various effects are not independent from each other, you end up with highly nonlinear equations that can result in chaotic behavior. As you may know, it really doesn’t take much nonlinearity to give chaos. The problem with chaos is that if you want to validate your model with past data (the only way it can be done with climate), you need to know the initial conditions with infinite accuracy to account for future evolution. On the other hand, over short time frames (a century is quite short), and small enough perturbations (both changes in solar and GHG emissions vary relatively little), you can linearize the equations and possibly get meaningful results.

BTW, I can’t understand the thing about “doubling” CO2. Doubling is a huge perturbation. CO2 has NOT doubled yet, it has only changed by about 30% as far as I know over 100 years. I just can’t see how you can use models based on such a slow increase and apply doubling of one parameter, and be condident that the answer will be right!

Does anyone know of such simplified models ? The advantages of a simple model is that simulations are much easier to carry out. If the model is adequate, in my opinion, the chances of getting meaningful results are higher than with a very complex model.

(OT my thing with Roctest is a long, and a bit of an extraordinary story, that dates back a few years. Most of the people I dealt with are now gone. I still live in Montreal. Yes, I worked in “optics”, but am now trying to reinvent myself, since the field doesn’t seem to want me anymore. I have a lot of “free” time (not so free ’cause I ain’t got no pay!), of which I spend an unordinate amount reading these blogs!)

Francois

Of course it aint that simple, there are various other factors that effect it, and it is a complex coupled chaotic system, but if the primary mover is solar (which makes a whole helluva lot of sense) and CO2 is a minor influence than all of this hoopla can take backstage where it belongs. If a simple algorithm can take into account Solar output, CO2 concentrations and accurately predict (let’s say within .5 degrees) global temperature, then we can insert various numbers for CO2 (Doubling) and determine it’s effect on global climate. But even with that, being the chaotic system that it is, you will never be able to truly predict it very accurately. And until we get some sunspot data during an ice age it will be hard to determine if the algorithm can predict that.

OT, funny that being in Montreal though it is somewhat to be expected. When was it? I spend a fair amount of time in Montreal because of it. Stop by Bishop st. some time. I perform a magic trick on Pints off Guinness there.

On another odd coincidence I used to work in Optics and if your doing work in that area it’s likely there are other coincidences.

SM: I edited this a little since it was spliced.

I wish it were all that simple! The solar hypothesis is appealing, but I’ve also seen criticisms of it. Quite frankly I don’t know enough to make my own opinion of it. There is a simple physical basis, but do the numbers add on? Apart from Soon and Balliunas, are there other references?

Francois O.

(btw, and completely off-topic, Erik, I used to work with Roctest. small world…)

besides a simple algorithm, I wonder what happens to Global temperature anomalies trends when you subtract out solar Forcing. Me-thinks it’s going to have a tendency to flat line them, while a small (minute) trend might be left after, certainly nothing to get excited about.