Tuesday, December 19, 2006

Why Water Doesn't Matter. Much.



Okay, there is obviously more to the Greenhouse Effect than a lame correlation between carbon dioxide concentration and temperature rise. I have only just figured out exactly how it all works, and maybe it is written up somewhere on RealClimate for noobs like me, but I couldn’t find it, so here it is.

Let us begin with the Sun. It is more or less a black body heated to a high temperature, and sends all kinds of electromagnetic radiation out in all directions, some of which impacts the Earth, as shown in Figure 1.

The difference between the upper dotted line (sunlight at the top of the atmosphere) and the lower solid line (sunlight at the bottom of the atmosphere) is the first lot of energy we need to worry about. Part of it looks like it is scattered back into space (the general fact that the solid line is lower than the dotted line) and part of it goes into increasing the kinetic energy of various molecules floating around in the air (those are all the little dimples in the solid line). These molecules (mostly water) can then knock into other molecules and increase the general kinetic energy- that is, the temperature- of the air. The more scatterers there are in the air- dust, soot, water droplets, etc.- the more energy will be scattered away, and the more water vapour (mostly) there is, the more the atmosphere will be heated directly. But on average, the solid line should not change much over time.

Now, what happens to the solid line when it reaches the earth’s surface? Either it will be reflected, and zip back off into space, or it will be adsorbed. This will be very variable indeed, and will depend on where the clouds are (they count as surface), and where the snow is, etc. Nobody is at all sure how this balance between reflection and adsorption will respond to an increase in global temperature, but the famous precautionary principle suggests that it is likely to stay about the same.

The adsorbed energy heats the Earth’s surface. But because the whole thing has to balance to keep the Earth’s temperature the same, it has to go somewhere: and where it goes is the energy radiated by a black body heated to a not-terribly-high temperature, as shown in Figure 2.



The heavy green line is the theoretical curve for a black body at 255 K, and the narrower green line is observational data from an area of the Pacific ocean at about 290 K. Now you can see the bending signal of carbon dioxide! This is the rational basis for being fretty about carbon dioxide. If the dip caused by carbon dioxide gets bigger, the total area of the curve has to increase to balance the average energy coming in with the energy being radiated out. Let’s say the dip increases to where it takes up an extra 10% of the total area under the curve: the surface temperature then has to increase by a factor of approximately the fourth root of 1.1, an increase of about 6 K. 10% is of course a ruinously gloom and doom eyeballing estimate by me that probably requires a quintupling of carbon dioxide concentration, so people are worried about an increase rather less than that.
This is why I was (probably) wrong about water vapour: Water vapour, though in one way of looking at things is responsible for 90% of global warming, in another way is irrelevant, since this emission is happening in a 'window' where water hardly absorbs at all.

Note that this 6 K with a vast increase of carbon dioxide is an increase in average surface temperature; not air temperature, which will be bouncing around all the time in response to the energy actually absorbed by the atmosphere directly, and the balance between reflected and absorbed radiation. I think this is the basis for the quarrel between the RealClimate guys, who think average air temperature is a good global warming way to measure nevertheless, and Roger Pielske Jr., who favours something to do with the heat content of the oceans as a better way to see how this balance between heat adsorbed and heat radiated is working out in practice.

No comments: