Info

FIGURE 14.27 Calculated fraction of 550-nm light scattered upward ( ¡3) as a function of solar zenith angle (0) and particle radius (/¿in). The refractive index of the particles is 1.4 (adapted from Nemesure et al., 1995).

FIGURE 14.27 Calculated fraction of 550-nm light scattered upward ( ¡3) as a function of solar zenith angle (0) and particle radius (/¿in). The refractive index of the particles is 1.4 (adapted from Nemesure et al., 1995).

details. The important point is that the combination (aRII/RH) is a very sensitive function of RH, and indeed, this appears to be the most important aerosol parameter in direct forcing (Nemesure et al., 1995; Pilinis et al., 1995). Figure 14.28 shows one calculation of aRII/Rn as a function of RH for (NH4)2S04, which is a common form of sulfate in tropospheric aerosols (Nemesure et al., 1995). The effect of relative humidity is large, increasing from ~f.5 m2 per g of sulfate for dry particles to ~50 m2 per g at 97% RH at a wavelength of 600 nm. Finally, as expected from Mie theory and demonstrated in Fig. 14.28, the light scattering efficiency varies widely over the wavelength region of interest in the troposphere.

The total emissions of S02 due to anthropogenic processes are believed to be relatively well known (see Chapter 2.A.4). A great deal is also known about the processes oxidizing tropospheric S02 to sulfate (see Chapter 8). In brief, while OH oxidizes S02 in the gas phase, oxidation in the liquid phase of fogs and clouds is generally more important. In the latter case, 03 is a major oxidant at high pH values, but as the droplet becomes acidified, this slows down. Over most of the pH range characteristic of particles in the troposphere, H202 is an effective oxidant and hence is believed to be the major contributor to sulfate formation. S(IV) may also be oxidized in the marine boundary layer by HOC1 and HOBr (Vogt et al., 1996). This means that the fraction of S02 forming sulfate will depend on such factors as the availability of liquid phase in the form of fogs and clouds, as well as the availability of oxidants such as H202, which is itself a complex function of the VOC-NOx chemistry discussed throughout this book. In addition, a large contributor to the removal of sulfate is wet deposition, which, due to its spatial and temporal variability, results in a corresponding variability in the tropospheric lifetime of sulfate. Because of all of these factors, there is significant uncertainty

FIGURE 14.28 Calculated mass scattering efficiency term (aR"/R") as a function of wavelength for 0, 40, and 97% RH. Particle dry radius taken to be 0.096-/im (NH4)2S04 (adapted from Nemesure et al., 1995).

associated with both ysu|lalc and tsu|lalc, and hence in the column burden of sulfate used in such calculations.

Finally, the term (1 — A,.) in Eq. (U), where AK is the fraction of the earth's surface covered by clouds, contains the implicit assumption that the direct scattering by aerosol particles is only significant in cloud-free regions. It is not clear, however, that this is the case. For example, while Haywood et al. (f997a) and Haywood and Shine (1997) report only a small contribution (5%) to aerosol particle direct forcing in cloudy regions, Boucher and Anderson (1995) report a much larger effect, with 22% of the total arising from cloudy areas. Other studies fall in between these values (e.g., see Haywood and Ramaswamy, 1998). For example, one calculation using a 1-D model suggests that cirrus clouds above a layer of particles of (NH)4S04 can enhance the direct forcing at a solar zenith angle of 0° from —2.0 to —3.3 W m~2 for their assumed set of conditions (Liao and Seinfeld, 1998). This is due to the fact that the overlying cloud scatters the incoming beam so that the light incident on the aerosol layer is effectively at larger solar zenith angles than 0°, leading to increased upward scattering (see the dependence of the upscattered fraction, ¡3, on solar zenith angle due to enhanced forward Mie scattering by particles, Fig. 14.27). A similar enhancement is predicted for a cloud layer of 100-m thickness assumed to be at an altitude of 900 m and embedded in a layer of aerosol taken to be uniformly distributed from 0 to 5 km. However, for much thicker clouds (e.g., 1000 m), the direct aerosol forcing due to sulfate may be decreased due to partial blocking of the incoming solar light intensity by the cloud (Liao and Seinfeld, f998).

In short, the approach summarized in Eq. (U) provided a useful first approach to establishing that direct scattering of light by sulfate particles could be important in counterbalancing the expected warming due to the increased greenhouse gases. However, given the large number of variables that enter into each term in this equation and their spatial and temporal variability, the development and application of more sophisticated models are clearly needed. For some examples of such models, see Kiehl and Briegleb (1993), Taylor and Penner Q994), Cox et al. (1995), Mitchell et al. (1995), Boucher and Anderson (1995), Meehl et al. (1996), C. C. Chuang et al. (1997), Haywood and Shine (1997), Haywood et al. (1997a, 1997b), van Dorland et al. (1997), Tegen et al. (1997), Schult et al. (1997), Haywood and Ramaswamy (1998), and Penner et al. (1998).

For example, Fig. 14.29 shows one calculation of direct radiative forcing using a global climate model, GCM (Penner et al., 1998). Due to the preponderance of anthropogenic S02 emissions in the Northern Hemisphere, the direct radiative forcing due to sulfate

FIGURE 14.28 Calculated mass scattering efficiency term (aR"/R") as a function of wavelength for 0, 40, and 97% RH. Particle dry radius taken to be 0.096-/im (NH4)2S04 (adapted from Nemesure et al., 1995).

FIGURE 14.29 Calculated direct radiative forcing due to sulfate aerosol particles (adapted from Penner et al, 1998).

FIGURE 14.29 Calculated direct radiative forcing due to sulfate aerosol particles (adapted from Penner et al, 1998).

aerosol is also predicted to occur in the Northern Hemisphere, particularly over the eastern United States, central and eastern Europe, and southeastern Asia. This model predicts a global average direct sulfate aerosol forcing of —1.2 W m~2 in the Northern Hemisphere compared to only —0.26 W m~2 in the Southern Hemisphere, where less than 10% of the anthropogenic sulfur emissions occur. Haywood and Ramaswamy (1998) predict very similar average values for the direct radiative forcing by sulfate, —1.4 W m~2 in the Northern Hemisphere and —0.24 W m~2 in the Southern Hemisphere, although there are some quantitative differences in the specific geographical dependencies.

It should be noted that the magnitude of the predicted forcing is quite sensitive to treatment of relative humidity (RH) in the model because of the effects on particle size and optical properties (e.g., Haywood and Shine, 1995; Haywood and Ramaswamy, 1998; Ghan and Easter, 1998; Haywood et al., 1998a; Penner et al., 1998). For example, in the calculations by Penner et al. (1998), when the particle properties were held fixed at the values for 90% RH for 90-99% RH, the predicted direct radiative forcing for sulfate particles decreased from - 1.18 W m~2 to -0.88 W nU2 for the Northern Hemisphere and from — 0.81 to —0.55 W m~2 globally.

Volcanic eruptions provide one test of the relationship between light scattering by sulfate particles and the resulting change in temperature, since they generate large concentrations of sulfate aerosol in the lower stratosphere and upper troposphere. These aerosol particles cause tropospheric cooling by backscattering solar radiation out to space. In principle, they can also cause tropospheric and stratospheric warming by absorbing and reemitting the long-wavength terrestrial infrared (McCormick et al., 1995; Robock and Mao, 1995); calculations show that the warming increases substantially with particle size and should be sufficiently large to counterbalance the cooling from light scattering at area-weighted mean radii larger than ~2 /¿m (Lacis et al., 1992). However, since the sulfate particles formed by gas-to-particle conversion of S02 are sub-yu,m (e.g., see Fig. f2.28), the backscattering effect leading to cooling generally predominates in the troposphere. Model calculations predicted that surface cooling of ~0.5°C should have occurred after the Mount Pinatubo eruption due to the scattering of solar radiation back out to space, and indeed, this is what was observed (e.g., Hansen et al., 1992; Dutton and Christy, 1992; Minnis et al., 1993; Lacis and Mishchenko, 1995; Robock and Mao, 1995; McCormick et al., 1995; Parker et al., 1996; Saxena and Yu, 1998).

Figure 14.30, for example, shows the predicted and measured monthly surface temperatures obtained prior to and following the eruption, demonstrating excellent agreement between observations and the model; the observed stratospheric warming is also in excellent agreement with model calculations (e.g., Lacis and Mishchenko, 1995; Angell, 1997). Temperature trends on a smaller, regional scale have also been attributed to the effects of such volcanic eruptions, although obtaining statistically robust results is difficult in these cases due to natural variability (e.g., Saxena et al.,

Was this article helpful?

0 0

Post a comment