Okay - next question to our resident expert (TennTradition).
Sweet...can I get VN Guru status for that
Well, first I think that I should preface my answers with the fact that the foundation of what I know about GW science was taught to me by a climatologist who studies GW and contributes to the IPCC as a lead author. He believes in the science...and uses it / furthers it through his research. It is also worth noting that he didn't fully come on board until the last 5-10 years (can't remember how long), when the uncertainty in basis of the models became small enough that he had confidence in the predictions - until then he was in an "agnostic" camp.
I've tried to advance my understanding of GW science/issue by reading views of both sides, yet my foundation was set by a "believer" - but someone who I firmly view as an honest scientist.
1. The impact of carbon emissions on the global climate:
a. is it's role settled?
b. is the magnitude of its role settled?
a)is it's role settled?
By role, I assume that you mean whether or not carbon emissions have an effect. I personally believe this is settled science. The role of a gas in climate forcing is a matter of physics (adsorption of energy, band gaps, etc.) and is very well understood (but not so much by me - engineer here, not a physicist).
An interesting side note here is water as a GHG. People who oppose the idea of GW often point to the fact that water is a much "stronger" or "more prevalent" GHG. But, then some argue that CO2 isn't. Now, the percentage argument aside (because that has already been addressed but can be confusing), this seems like a silly argument to me. We can't easily measure the effect of water vapor on temperature because we're already at saturation. So, unlike CO2, for which we have direct empirical evidence, we have to rely on the physics of water molecules to tell us how strong of a GW gas water vapor is. Just a side note...to explain why the physics are important.
b) is the magnitude of its role settled?
The magnitude of the role is up for more debate. I think that we can get better here, but much of this issue (temperature effect) is settled. The issues are exactly what level of radiative forcing will we see from GHGs at specific concentrations, what will be the climatic sensitivity to that forcing, and the question of how good our current means of answering those questions are.
We currently use a combination of the physics determining global warming potential of gases to determine the radiative forcing of GHGs. This is generally related to part a...and the science is quite solid.
Climate sensitivity bears much more uncertainty. Climate sensitivity answers the question of how much temperature rise we will see given a specific radiative forcing from a GHG (that is, given a specific increase in the amount of the sun's energy that gets trapped in the earth's atmosphere by the GHGs). We can predict climate sensitivity, but we can also measure it. The problem with measurements, though, is in correlated effects of many variables. For example, we know that CO2 has risen from 280 ppm to 380 ppm, and we are pretty confident that during this period global mean temperature have increased 3 C (made up number). So, you could say then that the climatic sensitivity to CO2 is 0.03 C/ppm CO2. But, obviously there are many other factors that could have impacted temperature during this period, such as the largest potential source - solar radiative flux increase from solar output or orbital pattern - or perhaps contributions from other GHGs. Also, some things could have been counteracting the warming during this period, such as volcanic eruptions leading to cooling. Thus, all of these factors lead to uncertainty in ascribing a specific climatic sensitivity to a given radiative forcing. Many respectable people have spent a lot of hours trying to answer these questions. Through their work, the IPCC has been able to conclude that the temperature rise due to anthropogenic emissions of GHGs has been, I think, 1.8 degrees C since the pre-industrial era. They ascribe something like a 95% confidence to this number (the uncertainty generated by all the factors I described above). I would say this number is reasonable.
2. Accuracy of predictive models:
a. magnitude of GCC expected
b. duration/direction of GCC expected
c. impact of GCC
a)magnitude of GCC expected
This is a very interesting question, particularly coming from someone who understands markets and has studied economics. Because, at the heart of this question are the social sciences - with other sciences thrown in like oceanography, meteorology, atmospheric sciences, physics, etc.
There are detailed economic models that attempt to track what consumption will be, what energy portfolios will look like, what policies may be undertaken, etc. to identify what the level CO2 output will be. This is a tough question...but good people work on it in many countries - and they can (although I admit I don't understand fully how) put error bars on their projections.
If you accept these numbers and error bars, the next question is how the environment will respond to these outputs. What will the rate of uptake of CO2 by the oceans, the land, etc. be? How will it change with increasing CO2 concentrations and increasing temperatures? How fast will the uptake mechanisms remove the new CO2? This is helped out some by data measurements over the last 50 years. So, you can gain some confidence in this.
In general, I believe the projections for CO2 concentration. They don't take into account radical technological changes or world-wide economic collapse (or unexpected advances), but these things can't really be accounted for. I would say the science is pretty good here - although I don't fully understand the social science side.
b) duration/direction of GCC expected
I'm not exactly sure what you meant here, but maybe I addressed this elsewhere?
c) impact of GCC
I'm going to break this into two sections, temperature and impact of this temperature
c1) impact on temperature
I've already talked about the role of GGC on past temperature increase. More questions come in about predicting future temperature increases. Scientists model this...and ascribe uncertainty through monitoring the affect of outcomes such as temperature after altering inputs with specific error bars (known as sensitivity analysis). However, to do this, you must identify and quantify feedback mechanisms...that is, specific factors in our environment that respond to temperature increases in a manner than either increases temperature further (positive feedback) or leads to a decrease in temperature (negative feedback). An example of a positive feedback is the ocean - increased temperature causes more CO2 to be released from the oceans causing further temperature increases. An example of a negative feedback is cloud formation - increased temperature is believed to cause larger cloud formation, and clouds have, in general, the effect of reflecting more radiation back to space than they trap/adsorb, leading to a decrease in temperatures. There are many, many positive and negative feedbacks in the models. But, what if a very important one has been missed? The scientists can do a pretty good job of modeling past temperature responses (obtained through surface measurements over time), which helps them build confidence that they have captured most important feedback mechanisms. But, that is not guaranteed.
So, here, I will say that while the science isn't settled - I believe the error bars are small enough to believe the temperature predictions. I wouldn't argue with you if you wanted to widen the uncertainty in the numbers a bit, though.
c2) impacts of temperature increases on environment
This is the holy grail of climate science, IMO. I say this because I think it is the hardest part to model and if we could overcome these problems, we could prepare for possible GHG concentration effects at local levels and prepare for the effects locally (and more efficiently).
Some effects are easier than others. If we want to understand sea level rise (due to expansion), then this we can do pretty accurately. There is some uncertainty in the width of the thermal boundary layer at the ocean surface and heat transfer rates, but in general, this can be predicted well. A more complicated aspect is ocean rise due to ice sheet melt. However, this is understood fairly well. The slight problem is that we are not beginning to talk about more local effects (polar), so you get more uncertainty. However, it is uncertainty associated with temperature, which is somewhat lower - so you can have a fair amount of certainty in these predictions. This is where a fairly large portion of the sea level rise prediction comes from - so it places larger error bars on sea level rise. I would say that the science could get better here - but it isn't incredibly far off (I would say it is fair here).
Other predictions are much harder....such as what will be the change in precipitation from state to state, or country to country. Predicting precipitation variations is kind of like predicting weather (the difference here is that you are smoothing over some small time-scale local variations to get mean or average values). However, I really don't buy it that much. The scientists are doing as good as they can - but that isn't good enough and probably won't be. I wouldn't attach much certainty to this at all. The models currently predict that the American Southwest will slowly transition into a grassland, but I wouldn't be rushing out to buy dessert land on that basis.