cherdano, on 2012-November-19, 16:05, said:
This is not how climate science works.
Too right. One of the main things against climate science is the lack of proof of the level required for physical sciences. If we take the Mann modelling technique for example, it uses a well-established mathemtaical process but throws out the key statistical test used to decide of the resulting data is meaningful, replacing it with a different test. Why? Because using the standard statistic says that the data is not meaningful. In the same vein, if you take only Northern hemisphere temperature records included in the modelling proxies you would probably expect a large LIA and MWP since both of these are reported in several regions. And yet, the dominance of the Urals data flattens out the resulting data to nothing within this period. I understand the argument that the Southern hemisphere data removes these "local" effects, but taking only Northern hemisphere data too? Come on, even ardent supporters of AGW must be able to see something is wrong there.
cherdano, on 2012-November-19, 16:05, said:
Climate scientists make models trying to predict future climate. Any such model is a huge simplification. They have to use their judgment to determine which simplifications still yield valid results. This judgment isn't completely subjective - it can be tuned based on past data, the results can be compared with other models, etc. But there is still enough judgment involved that a smart well-trained scientist with idiotic judgment may get wrong results. His paper showing "If we make the following assumptions, an increase in CO2 emissions will result in global cooling." could still be worth publishing, because, well maybe its true that a model with these simplifications predicts global cooling. This isn't physics where you can make hundreds of experiments - there i s only one climate, and experiments on its reaction to large scale greenhouse emissions cannot be run very frequently.
The way the models are produced, with forcings and feedbacks, provides a very wide array of possibilities that match the past data record. One of the positive thing about the models is that the outputs are reasonably consistent even for quite different input criteria. However, you are right to point out that all of the models use a great deal of subjective input. This is a real problem in a branch of science that is reliant on a particular hypothesis being correct and goes some way to explaining why almost every model uses a positive feedback for cloud formation even whe there is very little science to back this up. As a mathematician I believe strongly in modelling as a valid and useful approach to problems such as this. However, it is clear that you need to keep a skeptical head when producing such models, making sure to check a range of possibilities in unknown inputs. Otherwise you run the considerable risk of ending up with a GIGO model.
cherdano, on 2012-November-19, 16:05, said:
Your argument is a bit like saying "10 weather forecasters predict rain tomorrow, one predicts sun, so we can't conclude anything". Or "10,000 doctors claim I should quit smoking, but 20 think it increases my health, so there is nothing I can do until about my health until I have understood why each of these 20 doctors is wrong".
A better analogy is that 10 weath forecasters predict a 100% chance of rain for the next week, one predicts a 90% chance of rain every day. It only takes one dry day for the models used by the 10 to be faulty, even if it is only a small discrepancy. Or that 10000 doctors claim that I have a 100% chance of getting cancer because I smoke and my only chance of surviving is to give up smoking and pay $20000 per month for treatment (of which the doctors themselves get a 10% kickback), whereas 20 notice that if I pay $20000 per month I will be bankrupt and homeless in 6 months and suggest that I try an alternative treatment that only costs $20 per month but will reduce the chance of getting a fatal cancer to 10%.
I know these are stupid examples. So are yours. That is characteristic of the arguments on both sides sometimes. The truth is that we have the technology already to deal with excess greenhouse gases in the atmosphere. If it turns out that the problem is extreme then they could be implemented quickly. Naturally the cost of doing this would be greater than taking the same action over a longer time frame starting now. In this respect it makes sense to make changes now which have little impact on the existing economies and standards of living, or are already beneficial. It does not make sense to take the drastic action that has at times been suggested by some ecology groups (turn off all coal-powered power stations, etc).
One further example of this is the 4 degree link posted by PassedOut. Question: at the current rate of warming, how long will it take for the world to warm by 4 degrees? Given that solutions are available, is this a realistic scenario? You may as well post a link on why a 20 degree cooler world must be avoided. One example from the introduction: "Scientists agree that countries current United Nations Framework Convention on Climate Change emission pledges and commitments would most likely result in 3.5 to 4°C warming." So everyone agrees and the science is settled. Except that 3 paragraphs later comes: "We are well aware of the uncertainty that surrounds these scenarios and we know that different scholars and studies sometimes disagree on the degree of risk." Errrr, so is the science settled or not? This is Wayne's point. Also, the very next sentence is: "But the fact that such scenarios cannot be discarded is sufficient to justify strengthening current climate change policies." Scientists also cannot rule out the scenario of aliens landing on Earth and starting an intergallactic war - does that mean we should strengthen our policies against an alien invasion? Nor of a large interstellar body crashing into Manhattan - perhaps we should proactively evacuate all the inhabitants to underground bunkers just to be on the safe side.
This is precisely the problem. For every piece of rubbish AI posts here there is at least one equally as misleading posted on the AGW side. It is a little like the advertising for the American presidential race. Separating out the pieces of science based on good maths and statistics is not an easy task and it is no surprise that some sections of society have started to wonder what to think. It is tough enough even when you have some knowledge of the methods involved. Let us make an agreement - any link posted in this thread is regarded as bogus until someone else posts a meaningful argument as to why the underlying science in the linked article is indisputable. My guess is that very few links will meet this criteria (ignoring fanboi arguments).