Leaving aside the provision of employment for modelers, I am far from convinced that the climate change models are of any use at all. As an example, we often hear the proposition that to fix climate change we should find a way to get carbon dioxide from the atmosphere, or from the gaseous effluent of power stations. This sounds simple. It is reasonably straightforward to absorb carbon dioxide: bubble the gas through a suitable base. Of course, the problem then comes down to, how do you get a suitable base? Calcium oxide is fine, except you broke down a carbonate at quite high temperatures to get it. Amines offer an easier route, but to collect a power station's output, regenerate your amine, and keep the carbon dioxide under control will require up to a third of the power from your power station. Not attractive. The next problem is, what to do with the carbon dioxide? Yes, some can be sunk into wells, preferably wet basaltic ones as this will fix the CO₂, and a small amount could be used as a chemical, say to make polycarbonates, but how many power stations do you think will be accounted for by that?
 
The problem for climate change is that we currently burn about 9 Gt of carbon per annum, which means we have to fix/use something like 33 Gt of CO₂ per annum just to break even, and breaking even is unlikely to fix this carbon problem. The problem is, CO₂ is not a very strong greenhouse gas, but it does stay around in the atmosphere for a considerable time. One point that nobody seems to make in public is that even if we stopped emitting CO₂ right now, the additional carbon we have already put in the atmosphere will remain for long enough to do a lot more damage. Everybody seems to behave as if we are in a rapid equilibrium, and that is not so. The Greenland ice sheet is the last relic of the last ice age. If we have created the net warming to melt so much per annum, that will keep going until the ice retreats to a position more resilient, at which point our climate will change significantly because we have a much different albedo over a large area. We cannot "fix" climate change by simply stopping the rate of increase of burning carbon; we have to actively reduce the total integrated amount, and not simply worry about the rate of increased production. I suggest that to fix the climate problem, assuming we see it as a problem, we would be better to put more effort into something with a stronger response than fixing CO₂.
 
In the previous post, I attempted (unsuccessfully!) to irritate some people relating to how climate change research is spent. When money becomes available for this, what happens? What I believe happens is that we see numerous proposals for funding to make more accurate measurements of something. My argument is, just supposing we do get more accurate data on, say, the methane output of some swamp, what good does that do? It provides employment for those measuring the output of the swamp, but then what? Certainly it will add more to the literature, but the scientific literature is hardly short of material. Enough such measurements will help models account for what has happened, perhaps, but the one thing I am less confident about is whether such models will be able to answer the question, "Exactly what will happen if we do X?" For example, suppose we decided to try to raise the albedo of the planet by reflecting more light to space, and did this in a region that would lower the temperature of cold fronts coming into Greenland, with the aim of increasing snow deposition over Greenland, how much light would we need to reflect and where should we reflect it? My argument is, until models can give an approximate answer to that sort of question, they are useless. And unless we do something like geo-engineering, we are doomed to have to accommodate the change, because nobody has suggested any alternative that has the capacity to solve the problem. We can wave our hands and "feel virtuous" for claiming that we are doing something, but unless the sum of the somethings solves the problem, it is a complete waste of effort. Worse than that, such acts consume resources that could be better used to accommodate what will come. The only value of a model is to inform us which actions will be sufficient, and so far they cannot do that.