Net Radiative Effect of Clouds on Climate
At the heart of the mystery of how clouds affect climatic change is that clouds both cool and heat the planet, even as their own properties are determined by the cooling and heating. The cooling properties are readily visible: the minute water or ice particles in clouds reflect between 30 and 60 percent of the sunlight that strikes them, giving them their bright, white appearance. (Large bodies of water such as lakes and oceans absorb more sunlight than they scatter and so appear dark.) A cloudless Earth would absorb nearly 20 percent more heat from the sun than the present Earth does. To be in radiation balance Earth would have to be warmer by about 12°C (22°F). Clouds cool the planet by reflecting sunlight back into space, much as they chill a summer's day at the beach.
The cooling effect is largely offset, however, by a blanketing effect: clouds reduce the amount of heat that radiates into space by absorbing the heat radiating from the surface and reradiating some of it back down. The process traps heat like a blanket and slows the rate at which the surface can cool. The blanketing effect warms Earth's surface by some 7°C (13°F). Thus the net effect of clouds on the climate is to cool the surface by about 5°C (9°F). One can calculate the higher surface temperatures that must result from the buildup of greenhouse gases (and the consequent slowing of heat radiation) provided nothing else changes. But what happens to the radiation balance if, as part of the climatic response, the clouds themselves change?
Clouds are also involved in another heat exchange process where water evaporating from and cooling Earth's surface heats the atmosphere when rain and snow form in clouds. This water cycle is also important because of its affect on our water supply and agriculture.
If the cooling effect of clouds increases more than the heating effect does, the clouds would reduce the magnitude of the eventual warming but speed its arrival. The same result could come about if both effects decrease, but the cooling decreases less than the heating does. If the cooling increases less (or decreases more) than the heating, the cloud changes would boost the magnitude of eventual warming but delay its arrival. (It is also possible for the two effects to go in opposite directions, which would give rise to outcomes similar to the ones mentioned, but more intense.) In any event, what matters is only the difference between the cooling and the heating effects of clouds.
For a more detailed and technical discussion, see Rossow and Zhang 1995 and the references therein.