CCN and Their Enhancements

Aerosols can act as CCN and modify as a result cloud microphysics with potential impacts on cloud macrophysical properties (e.g., cloud structure, frequency, or lifetime) and the hydrological cycle (e.g., precipitation intensity, frequency, or distribution; cf. Stratmann et al. and Ayers and Levin, both this volume). Thus, there is a need to quantify the number of available aerosol particles that can serve as nuclei. In the context of a changing climate, there is particular interest in anthropogenic CCNs.

Aerosol particles, acting as CCN, allow atmospheric water vapor to condense and form cloud droplets. This condensation occurs preferably on larger particle sizes, because lower supersaturations are required to overcome their surface curvature effect. Thus, only the larger particle sizes are of interest

Anthropogenic aerosols o, u 0.082

Anthropogenic aerosols o, u 0.082

0.0000

0.2500

0.401

0.5000

Figure 3.8 Solar spectral variations for single-scattering properties of anthropogenic aerosols. Annual global fields for the column-integrated properties of extinction (AODa), single-scattering albedo (ffl0A), and asymmetry factor (gA) are presented at an UV (380 nm), a VISIBLE (550 nm) and two near-IR (1050 nm, 1585 nm) wavelengths.

0.570

0.0000

0.2500

0.473

0.401

0.5000

0.570

0.7500

Figure 3.8 Solar spectral variations for single-scattering properties of anthropogenic aerosols. Annual global fields for the column-integrated properties of extinction (AODa), single-scattering albedo (ffl0A), and asymmetry factor (gA) are presented at an UV (380 nm), a VISIBLE (550 nm) and two near-IR (1050 nm, 1585 nm) wavelengths.

(here, as well as for radiative transfer impacts). Potential CCN are all aerosol particles of the coarse size mode and the larger sizes of the accumulation size mode. Differences in the hygroscopic particle properties are a complicating factor. However, recent studies (see discussion in Part 2) have concluded that hygroscopic particle properties cluster around characteristic values over land and ocean. Assuming this simplification, the critical size, above which particles can serve as CCN, is primarily a function of the water vapor supersaturation. Typical values for supersaturation are near 0.1% and up to 0.5% for more con-vective cloud systems. An analytical formula (Rose et al. 2008; see also below, Equation 3.3) places the critical radii at 80 nm (over land) and 30 nm (over oceans) for a supersaturation of 0.1% and at 60 nm (over land) and 20 nm (over oceans) for a supersaturation of 0.5%. Thus, over land as much as 50% of the pollution or biomass (accumulation size) aerosol may be too small to serve as CCN, whereas over the ocean almost all accumulation (and of course coarse) particle sizes can be potential CCNs.

Aside from data on hygroscopicity and supersaturation, estimates for CCN require information on size and (local) aerosol amount. This information is supplied by concepts and data from the new aerosol climatology. Aerosol amount is based on global AOD maps, where sun photometer data are merged onto a modeling background. For the necessary vertical distribution of the AOD, monthly average data from global model simulations were applied. As part of g

AeroCom model evaluations (Giubert, pers. comm.), general agreement has been demonstrated between simulated aerosol vertical distributions and lidar profiles. However, given the diversity in modeling and a growing database by active remote sensing from space, it is recognized that the use of model data at this stage is a pragmatic choice to satisfy data needs. At a later stage, data on aerosol vertical distribution will certainly be better constrained by active remote-sensing data from ground and space. Particle size is based on a bimodal, log-normal distribution, and a distinction is made between coarse and accumulation size modes. Apportionment of AOD to each mode (AODa, AODc) is tied to the mid-visible AOD spectral dependence, combining the spectral insensitivity of the coarse mode with a prescribed spectral dependence for the accumulation mode. The coarse mode composition (either dust or sea salt) as well as the dust size are defined by the mid-visible ro0. More specifically, the coarse mode assumes a log-normal distribution with a fixed distribution width (standard deviation 2.0). The assumed mode radii are 0.75 pm for sea salt and 0.375 pm for dust. It should be noted, however, that larger dust sizes are successively chosen (0.75, 1.5, or 3.0 pm), if (small) particle absorption alone is unlikely to explain locally the low ro0 of the climatology. The smaller accumulation mode also assumes a log-normal size distribution with a fixed (though narrower) distribution width (standard deviation 1.7). In conjunction with the prescribed AnP (completely dry: 2.2; completely wet: 1.6) the mode radius is defined to lie between 0.085 pm under completely dry and 0.135 pm under completely wet conditions (where low cloud cover of a cloud climatology is applied to define wetness).

Three types of CCN concentrations are considered. In the first scenario, all aerosols of the coarse and accumulation modes are considered. In the second scenario, supersaturations of up to 0.5% are permitted, which requires that the log-normal distribution of the accumulation mode needed to be truncated from sizes smaller than 60 nm over land and smaller than 20 nm over oceans, since these particles were too small to be activated. In the third scenario, the maximum supersaturation was set to 0.1%, with cutoff sizes at 80 nm over land and 30 nm over oceans. Simulated CCN concentrations at about 1 km above the ocean or land surface are displayed in Figure 3.9 (in log10 space) separately for total, natural, and anthropogenic aerosol.

The CCN concentrations of Figure 3.9 are given for a logarithmic scale and annual averages. Monthly CCN fields for total (natural and anthropogenic) aerosol at 0.1% and 0.5% supersaturations indicate seasonal variations (e.g., increases during the tropical biomass burning season) and demonstrate that CCN concentrations increase (on a global average basis) by about 30%, when relaxing the supersaturation from 0.1% to 0.5%. It should be noted that CCN concentrations of the accumulation mode are about one order of magnitude larger than CCN concentrations of the coarse mode. Thus, it is not surprising that monthly maps for the CCN anthropogenic fraction resemble the monthly maps for the anthropogenic fraction for AODA of Figure 3.7.

1 km concentration All 0.5% 0.1%

? fP

ctt

ct5

ct1

7.B41

7.479

7.364

i '

iMunrm

cnt

cn5

cn1

7.136

7.134

7.025

--

- --a

1ยป

r

cat

ca5

ca1

7.091

7.089

6.970

_

ST*?*

Figure 3.9 Annual average global maps for CCN concentrations without cutoffs to sizes of the accumulation mode (left), at 0.5% supersaturation (center), and at 0.1% supersaturation (right). Concentrations are displayed (in log10 m 3) separately for total aerosols (top), natural aerosols (center), and anthropogenic aerosols (bottom).

Figure 3.9 Annual average global maps for CCN concentrations without cutoffs to sizes of the accumulation mode (left), at 0.5% supersaturation (center), and at 0.1% supersaturation (right). Concentrations are displayed (in log10 m 3) separately for total aerosols (top), natural aerosols (center), and anthropogenic aerosols (bottom).

CCN concentration changes with altitude. Thus, to obtain information on cloud development at higher altitudes, CCN concentrations were determined at two additional altitudes: 3 km and 8 km. Simulated CCN concentrations for a 0.1% supersaturation at these three altitudes are compared in Figure 3.10 (in log10 space), again separately for natural, anthropogenic, and total (anthropogenic and natural) aerosol.

The simulated total (natural and anthropogenic) CCN concentrations are in general agreement with Glomap simulations (Spracklen et al. 2008). Their global annual average at the surface and for a 0.2% supersaturation is about twice as large as the estimate of this study for 1 km altitude and a 0.1% supersaturation. There is also agreement on the seasonal cycle, as July CCN concentrations (northern hemispheric summer) are higher than for December. Most maxima match (e.g., industrial regions). The largest differences are Glomap-simulated CCN sinks in the ITCZ, which are likely caused by cloud-processing in the Glomap model.

Of particular interest are the CCN enhancement factors, defined by the ratio between anthropogenic and natural concentration. Since the annual ratios resemble each other at different supersaturations and altitudes, enhancement factors are presented on a monthly basis in Figure 3.11 for the most interesting case of low-level water clouds: lower altitude and 0.1% supersaturation.

CCN at 0.1% supersaturation Total Natural Anthropogenic

CCN at 0.1% supersaturation Total Natural Anthropogenic

Figure 3.10 Annual average global maps for CCN concentrations at 0.1% supersaturation at different altitudes: 1 km above the ground, at 3 km, and at 8 km. Concentrations (in log10 m 3) are shown for total, natural, and for anthropogenic aerosols.

Figure 3.10 Annual average global maps for CCN concentrations at 0.1% supersaturation at different altitudes: 1 km above the ground, at 3 km, and at 8 km. Concentrations (in log10 m 3) are shown for total, natural, and for anthropogenic aerosols.

The monthly maps in Figure 3.11 illustrate that anthropogenic enhancements occur primarily in the Northern Hemisphere, predominantly near industrial areas. Anthropogenic CCN enhancements are greater during the winter season, at which time they significantly impact the Arctic.

0 0

Post a comment