distribution (Pandis et al., 1994). Other problems, including those involving aerosol-cloud interactions, require ~100 size categories to accurately represent the indirect effect of particles on cloud properties (Russell et al., 1999). While the simplified models are known to represent measured size distributions incompletely, the uncertainties associated with these simplifications can easily be quantified. GCMs employing this approach have not yet had the computational power to evaluate the sensitivity of aerosol radiative forcing to their simplified treatments of aerosol size distributions.
Initial models of particle evolution assumed that the composition of particles consists of a single internal mixture, both for computational simplicity and for lack of contradictory measurements (Warren and Seinfeld, 1985). In the last decade, chemical composition has been discretized in a fashion similar to the treatment of particle size, typically by “lumping” particles with similar chemical compositions into a few categories. This approach has been used to represent the externally mixed nature of aerosol particles, typically including categories such as “sulfate,” “sea salt,” “dust,” and “carbonaceous” (Jacobson et al., 1994; Pandis et al., 1994; Russell and Seinfeld, 1998; Jacobson, 2001; Koch, 2001; Garrett et al., 2003). The small number of categories has enabled their inclusion in GCM simulations. However, this does not reflect the variety of mixtures actually present in the troposphere (Murphy et al., 1997). The complexity of particle structures, their heterogeneities, and their mixing states (Russell et al., 2002) will have to be addressed to represent their hygroscopic and optical properties. For example, a small amount of absorbing organic compounds mixed in with sea salt aerosol can be sufficient to change the associated radiative forcing from negative to positive, especially over low albedo surfaces such as clouds, ice, and coastal areas (Randles et al., 2004).
Aerosols interact with clouds in several ways that can affect the climate system, in particular by acting as cloud condensation nuclei (CCN) and ice nuclei. These interactions are generally referred to as the indirect effects of aerosols. Table 2-2 summarizes the various aerosol indirect effects. As shown in the table, aerosols can lead to both warming (positive forcing) and cooling (negative forcing), complicating the analysis of their net effect.
Aerosols have several indirect effects on warm stratiform clouds. The Twomey effect, also known as the first indirect aerosol effect, refers to the enhanced reflection of solar radiation due to more but smaller cloud droplets in a cloud whose liquid water content remains constant (Twomey, 1959). The IPCC Third Assessment Report concluded that the first indirect aerosol effect amounts to 0 to −2 W m−2 in the global mean (IPCC, 2001). In addition, more but smaller cloud droplets reduce the precipitation effi-