FIGURE 14.40 Fraction (F) of aerosol particles that are activated to form cloud droplets as a function of the total number of particles (JV101). The horizontal line represents the 50th percentile for 10 sets of aircraft measurements. The 5th, 25th, 75th, and 95th percentiles are also shown (adapted from Gillani et al., 1995).

or as much as an order of magnitude smaller, depending on a number of factors, the most important of which are the particle sizes, the total particle concentration, and the local cooling rate in the cloud (Gillani et al., 1995).

An increase in aerosol particles that can act as CCN can increase the number of cloud droplets and their size distribution, both of which can affect the light scattering properties of clouds and hence climate. We first briefly discuss the effects of clouds on climate and then the potential impacts of anthropogenic aerosols on the formation and properties of clouds.

Clouds and global climate. Clouds in the troposphere interact with both solar and terrestrial radiation in complex ways, and either warming or cooling can result (e.g., see Ramanathan et al., 1989; Ramaswamy and Ramanathan, 1989; Fouquart et al., 1990; Harrison et al., 1990; Liou, 1992; King, 1993; Hartmann, 1993; Rossow and Zhang, 1995; Ramanathan, 1995; Crutzen and Ramanathan, 1996; and Baker, 1997). Thus, marine stratiform clouds found in the boundary layer backscatter solar radiation, leading to negative radiative forcing and a net cooling. Although such clouds also absorb terrestrial infrared radiation, they emit at about the same temperature as the earth's surface. As a result, as discussed for the greenhouse gases, there is little net effect (see Section A.2). However, cirrus clouds and deep convective cloud anvils found near the tropopause emit long-wavelength infrared out to space (Fig. f4.2c) at the colder temperatures characteristic of this region of the atmosphere. Because this energy emission is at lower temperatures, the net infrared emission out to space when they are present is smaller, leading to a positive radiative forcing, i.e., to warming (Twomey, 1991). As a result, the net radiative forcing due to low clouds over the oceans is generally negative, whereas it is positive over some continental regions with high clouds. On a global basis, the mean net effect is negative, ~ —20 W m~2 (Baker, 1997). For reviews of the relationship between clouds and climate, and anthropogenic effects on them, see Hobbs (1993a, 1993b), King (1993), Hartmann (1993), Andreae (1995), Schwartz and Slingo (1996), Schwartz (1996), and Baker (1997).

Since some aerosol particles, which may be solids or liquids, e.g., H2S04, can serve as cloud condensation nuclei (CCN), increased particle emissions from anthropogenic processes have the potential for increasing the number of CCN. The concentration of droplets in a cloud is therefore expected to increase, although not necessarily in a linear fashion, with the increased concentrations of aerosol particles. The formation of a larger number of droplets for a given liquid water content will lead to each droplet being smaller, i.e., shift the size distribution to smaller droplets. This effect increases the cloud albedo and hence can contribute to global climate change (Twomey, 1974, 1977a,b, 1991; Twomey et al., 1984). Evidence for these effects is discussed shortly. Finally, the shift to smaller drop sizes may decrease the precipitation rate from clouds, increasing their lifetimes and hence the average amount of cloud cover (Albrecht, 1989; Lohmann and Feichter, 1997). This is also expected to have a significant effect on global climate.

It should be noted that while aerosol particles affect clouds by serving as CCN for cloud formation, the reverse is also true, i.e., clouds also affect the formation and size distribution of aerosol particles. For example, the oxidation of S02 to sulfate in clouds generates larger particles whose light scattering cross sections are larger than for smaller particles formed by gas-phase oxidation processes (e.g., see Lelieveld and Heintzenberg, 1992). Aerosol particles incorporated into a cloud droplet will reappear as particles when the cloud evaporates. However, if new aerosol constituents are formed by in-cloud oxidation (e.g., of S02 to sulfate), the size of the resulting particle will be larger than the original particle. As discussed earlier, this is a potentially important process for growing aerosol particles, which are too small to serve as CCN at low values of Smax (e.g., for marine stratus), into sufficiently large sizes (i.e., to the right of the peak in Fig. 14.38) that they can act as CCN under the appropriate conditions of supersaturation (e.g., see Hegg, 1990; and Kaufman and Tanre, 1994).

For example, as seen in Fig. 14.41, aerosol particle number size distributions in the clean marine boundary layer outside of clouds are often observed to have a bimodal distribution. The larger mode above 0.1 /¿m

FIGURE 14.40 Fraction (F) of aerosol particles that are activated to form cloud droplets as a function of the total number of particles (JV101). The horizontal line represents the 50th percentile for 10 sets of aircraft measurements. The 5th, 25th, 75th, and 95th percentiles are also shown (adapted from Gillani et al., 1995).

FIGURE 14-41 Typical particle size number distributions for marine aerosols outside of clouds where the total aerosol number concentration was < 500 cm~3 (adapted from Anderson et al., 1994).

has been attributed to aerosol particles that have been "cloud processed" (Hoppel et al., 1986; Hoppel and Frick, 1990; Anderson et al., 1994). That is, these particles served as CCN upon which clouds formed, followed by aqueous-phase reactions and evaporation of the cloud droplets to leave larger particles. Indeed, in some studies (e.g., Van Dingenen et al., 1995), it is assumed that particles in this mode can be taken as a measure of the CCN that were available for cloud formation in the prior cloud event. As discussed in Chapter 9.A.2, such aqueous-phase processes in the atmosphere are also believed to lead to two peaks in accumulation mode particles in urban areas.

The albedo (R) of a thick, boundary layer cloud that does not absorb solar radiation over a surface with zero albedo can be approximated (Twomey, 1991; Schwartz and Slingo, 1996; Baker, 1997) by

Albedo = R

Was this article helpful?

0 0

Post a comment