Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

Figure 5.5: Flow chart showing the processes linking aerosol emissions or production with changes in cloud optical depth and radiative forcing. Bars indicate functional dependence of the quantity on top of the bar to that under the bar. Symbols: CCN (Cloud conden-sation nuclei); CDNC (Cloud droplet number concentration); IN (Ice nuclei); IP (Ice particles); OD (Optical depth); HC (Hydrometeor concentration); A (Albedo); fc (Cloud fraction); c (Cloud optical depth); F (Radiative forcing).

5.3 Indirect Forcing Associated with Aerosols

5.3.1 Introduction

Indirect forcing by aerosols is broadly defined as the overall process by which aerosols perturb the Earth-atmosphere radiation balance by modulation of cloud albedo and cloud amount. It can be viewed as a series of processes linking various intermediate variables such as aerosol mass, cloud condensation nuclei (CCN) concentration, ice nuclei (IN) concentration, water phase partitioning, cloud optical depth, etc., which connect emissions of aerosols (or their precursors) to the top of the atmosphere radiative forcing due to clouds. A schematic of the processes involved in indirect forcing from this perspective is shown in Figure 5.5. Rather than attempt to discuss fully all of the processes shown in Figure 5.5, we concentrate here on a selected suite of linkages, selected either because significant progress towards quantification has been made in the last five years, or because they are vitally important. However, before delving into these relationships, we present a brief review of the observational evidence for indirect forcing.

5.3.2 Observational Support for Indirect Forcing

Observational support for indirect forcing by aerosols derives from several sources. Considering first remote sensing, satellite studies of clouds near regions of high SO2 emissions have shown that polluted clouds have higher reflectivity on average than background clouds (Kuang and Yung, 2000). A study by Han et al. (1998a) has shown that satellite-retrieved column cloud drop concentrations in low-level clouds increase substantially from marine to continental clouds. They are also high in tropical areas where biomass burning is prevalent. Wetzel and Stowe (1999) showed that there is a statistically significant correlation of aerosol optical depth with cloud drop effective radius(reff) (negative correlation) and of aerosol optical depth with cloud optical depths (positive correlation) for clouds with optical depths less than 15. Han et al. (1998b), analysing ISCCP data, found an expected increase in cloud albedo with decreasing droplet size for all optically thick clouds but an unexpected decrease in albedo with decreasing droplet size in optically thinner clouds (tc<15) over marine locations. This latter relationship may arise because of the modulation of the liquid-water path by cloud dynamics associated with absorption of solar radiation (Boers and Mitchell, 1994) but may also arise from the generally large spatial scale of some satellite retrievals which can yield misleading correlations. For example, Szczodrak et al. (2001), using 1 km resolution AVHRR data, do not see the increase in liquid-water path (LWP) with increasing effective radius for all clouds seen by Han et al. (1998b), who utilised 4 km resolution pixels. In any case, a relationship similar to that found by Han et al. (1998b) was found in the model of Lohmann et al. (1999b,c) and that model supports the finding of a significant indirect forcing with increases in aerosol concentrations (Lohmann et al., 2000). Further evidence for an indirect forcing associated with increases in aerosol concentrations comes from the study by Nakajima et al. (2001). They found increases in cloud albedo, decreases in cloud droplet reff, and increases in cloud droplet number associated with increases in aerosol column number concentration.

Continues on next page



Other reports in this collection