Tucson, AZ 85730
Introduction
In predicting the global warming due to a doubling of CO2, it is important not to only evaluate the net effect of all the known feedback mechanisms, but to estimate the sensitivity to each. In other words, the partial derivatives as well as the total derivatives should be estimated. For instance, we might conclude that
Tbar/ ln[CO2] = +1.5K
Tbar/ ln[H2O(vap)] = +12.5K
indicating that the climate is more sensitive to changes in water vapor which may occur as a result of CO2 greenhouse warming, than to the CO2 induced warming itself.
Precipitation
In a simple model, water vapor is injected at a rate proportional to the saturation vapor pressure, and rained out at an equal rate. The rate of precipitation (according to Berry) is proportional to the cube of LWC, since droplet coalescence is a highly nonlinear process. (In this sense, it is not the average LWC that matters for precipitation, but the extreme values in a cloud). In order for relative humidity to remain constant, the LWC must be proportional to the cube root of the saturation vapor pressure, and it is difficult to explain why this should be true. Consequently, there needs to be a better justification for making climate predictions which assume constant relative humidity.
Infrared Radiation
The treatment of clouds as black bodies is only an approximation. In fact, the extinction and absorption properties of cloud drops are quite sensitive to droplet size for the sizes typically found in clouds. Detailed calculations of the infrared "seeing" distances in clouds indicate that continental (polluted) air masses produce clouds with significantly smaller seeing distances than do maritime air masses (Figure 22.1). This effectively means that cloud top cooling will occur over a smaller depth in continental clouds, producing a more intense response (smaller time constant).
Shortwave Radiation
It is important to quantify the influence of cloud absorption. Figure 22.2 shows isolines of reflectance in the absorption/droplet concentration plane. Reflectance strongly depends on droplet concentration, but only weakly on absorption until the absorption becomes large (which is one reason that cloud absorption is so poorly determined at present). The important point is that reflectance can either increase or decrease with increasing droplet concentration depending on the way in which absorption increases with droplet concentration.
Shiptrack experiments help to quantify the effect of air pollution on clouds. Shiptracks are found to be twice as bright as cloud surrounding the tracks, implying a factor of 8 increase in droplet concentration, and consequently a halving of mean radius in order to conserve liquid water content. Results by Platnick (1991) add weight to this conclusion.
Three scenarios were considered for particles in a CO2 doubled atmosphere: (1) the aerosol loading would double, (2) the aerosol concentration would remain constant, but aerosol mass would increase, and (3) the coagulating mass flux would double. These three scenarios lead to a spread in climatic responses, (1) almost doubling the droplet concentration, (2) increasing it by only 20%, and (3) being in between. The climate sensitivities (see introduction) are, respectively, -2.8K, -0.7K and -1.5K. The point is that these sensitivities are as big as the direct CO2 doubling effect, so that our uncertainty about which scenario is most realistic has important implications for our global change predictions.
Conclusion
The approximate treatment of precipitation and infrared transmission in clouds together with uncertainties in the effect of a warmer climate on the global aerosol loading introduce potentially significant sources of error into current global change predictions.
