Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
ELECTRICAL STRUCTURE FROM 0 TO 30 KILOMETERS 168 AEROSOLS IN THE LOWER ATMOSPHERE Within the portion of the atmosphere under consideration in this chapter there are three primary regions of interest to atmospheric electricity: the boundary layer (approximately the first 3 to 5 km), the remaining portion of the troposphere, and the stratosphere. The characterizing aerosol parameters of central importance here are concentration, size distribution, and vertical structure. In addition, variability of the aerosol in each of these regions must be recognized. Such variations may influence electrical parameters and significantly detract from the apparent repeatability of various atmospheric electrical measurements. The boundary layer is generally thought of as a relatively well-mixed region capped on the upper side by a temperature inversion. Since mixing across the inversion tends to be inhibited, a potential exists for the accumulation of aerosols within the boundary layer if a significant source is present. The buildup of aerosols is frequently of sufficient magnitude to cause a notable reduction in visibility. Even when there is no apparent loss in visibility as observed from the surface, the upper boundary may still be apparent even to an airline passenger at the moment when the aircraft passes through inversion. The buildup of aerosols in the boundary layer and the capping effect of the inversion have been observed in a more quantitative sense by airborne lidar. The ion concentration and conductivity can be greatly affected by the aerosol buildup in the boundary layer and a dramatic change in these quantities is often observed at or near the altitude of the defining inversion. The size distribution of aerosols near the surface of the Earth has frequently been approximated with a powerlaw function (Junge, 1963). However, it has more recently been suggested that the near-surface aerosol is actually made up of two or three size modes that when added together approximate a power-law function over a limited range of sizes (Willeke and Whitby, 1975). This interpretation of the size distribution seems to better reflect the physical processes affecting the aerosol concentration in the atmosphere. Some examples of typical size distributions for a variety of conditions and locations can be found in the work of Willeke and Whitby (1975) and Patterson and Gillette (1977). The aerosol mixing ratio profile (and usually the concentration profile) typically show a relative minimum in the upper troposphere for particle radii greater than about 0.1 Âµm. In contrast the condensation nuclei (cn) profile (corresponding to particle radii of about 0.01 Âµm) is more or less constant throughout the entire troposphere above the boundary layer. Near the surface of the Earth the cn concentration may be relatively high owing to local contamination (104 per cm3 or more), but above the boundary layer the concentration ranges from about 100 to 103 per cm3 with a global average of approximately 300 to 500 per cm3 (Rosen et al., 1978a, 1978b). The size distribution of aerosols in this region of the atmosphere can usually be approximated by a power-law function between about 0.01- and 10-Âµm particle diameter (Junge, 1963). Below the minimum size of 0.01 Âµm there are relatively few particles owing to coagulation, and above 10 Âµm the particle concentration drops quickly from sedimentation effects. Thus limits of the power-law distribution must always be specified. The character of upper tropospheric aerosols can be temporarily disturbed by volcanic eruptions, forest fires, biomass burning, and large dust storms. In addition periodic annual variations of concentration have also been observed (Hofmann et al., 1975). Another important temporal variation of tropospherical aerosols is associated with the so-called arctic haze events. New evidence suggests that these events are characterized by high particle loading throughout a large portion of the troposphere. The impact of this extensive aerosol loading on atmospheric electrical parameters is yet to be determined. Hogan and Mohnen (1979) reported the results of a global survey of aerosols in the troposphere and lower stratosphere. They found that the concentrations were more or less symmetrically distributed about the Earth. Measurements of this type could provide the basis for extrapolating local or isolated observations to characteristic worldwide values. The morphology of stratospheric aerosols is dominated by a persistent structure frequently referred to as the 20- km sulfate layer, or Junge layer. It is now known that the character of this layer is highly affected by large volcanic eruptions. For several years prior to 1980, stratospheric aerosols were in a quasi-steady-state condition, not being under the influence of any significant recent volcanic eruptions. During that period the size distribution appeared to be consistent with a single mode log-normal distribution (Pinnick et al., 1976), although other types of single-mode distribution were also employed (Russell et al., 1981) with similar results. The composition was thought to be primarily sulfuric acid droplets. The appearance of the 20-km layer is not evident in the vertical profile of all particle size ranges. The cn profile, for example, which is representative of particles with sizes in the neighborhood of 0.01-Âµm radius usually shows a dramatic drop in concentration above the tropopause and no relative maximum at the altitude of the stratospheric aerosol layer. This would be consistent
ELECTRICAL STRUCTURE FROM 0 TO 30 KILOMETERS 169 with a tropospheric cn source and a vertical profile dictated by diffusion and coagulation (Rosen et al., 1978a). After a large volcanic eruption the size distribution may be greatly disturbed and highly altitude dependent. Following the eruption of El Chichon in April 1982, Hofmann and Rosen (1983a, 1983b) reported a size distribution that could be approximated by a sum of multiple log-normal distributions, each with a different mode. The source of the smallest particle mode ( ~ 0.01 Âµm) appeared to be associated with the homogeneous condensation of highly saturated sulfuric acid vapors. This production ceased a few months after the eruption, and consequently the mode disappeared. A midsize mode ( ~ 0.1 Âµm) was also observed, which may have resulted from coagulation and growth by condensation of the initial particles formed by the homogeneous condensation. Another mode near 1-2Âµm diameter was also observed that may have formed from the condensation of sulfuric acid vapors onto the solid silicate ash particles as well as on particles already present in the stratosphere. Further assessments of the aerosol injected by the El Chichon eruption have been described in a collection of papers introduced by Pollack et al. (1983). An unexpected influence of the El Chichon eruption was the enhancement of an annually appearing cn layer near 30 km altitude. Although the phenomenon was observed in previous years (Rosen and Hofmann, 1983) the concentration of cn associated with these event layers was enhanced by at least 2 orders of magnitude. Even though the sizes of the particles were quite small (~ 0.01-Âµm diameter), the concentration was large enough to measurably affect the ambient ion concentration and conductivity (Gringel et al., 1984). The event particles are thought to have formed in polar regions from highly supersaturated sulfuric acid vapors. The affect of this phenomenon on the various aspects of atmospheric electricity at high latitude has not been assessed at the time of this writing. Of particular interest would be the influence of the highly supersaturated vapors on the ambient ion mass (and therefore mobility). Careful measurements at appropriate locations may provide an unexpected means of finding supporting experimental evidence for some of the various models of ion composition (and mass) that have been proposed (Arnold, 1983; Arnold and BÃ¼hrke, 1984). Figures 12.2 to 12.4 illustrate several of the characteristics of atmospheric aerosols that have been discussed above. The long-term influence of volcanic aerosols on the stratosphere are illustrated in Figure 12.2, which compares the quasi-steady-state period with the disturbed conditions still observed some 18 months after the eruption of El Chichon. Note also the presence of the boundary layer near the surface (as evidenced by a sharp drop in concentration) and a relative minimum in the aerosol concentration occurring in the upper troposphere. Figure 12.2 A comparison of aerosol profiles (particle diameter greater than 0.30 Âµm) for the quasi-steady-state period (September 27, 1978) with the decay period following the April 1982 eruption of El Chichon. An example of a normal and disturbed cn profile is illustrated in Figure 12.3. A significant cn event layer is evident at about 30 km altitude. Both profiles show a relatively large drop in the cn concentration just above the surface, a relatively constant mixing ratio throughout most of the troposphere, and a noticeable drop in concentration near the tropopause. An example of the influence of volcanic eruptions on the size distributions of stratospheric aerosols is illustrated in Figure 12.4. As previously discussed, the size distribution during the quasi-steady-state period could be described quite well by a single-mode log-normal distribution. At the time of this writing some 18 months after the eruption of El Chichon, the size distribution appears still to be quite disturbed.