Although IPCC (1994) identified nitrate aerosol as a significant anthropogenic source of aerosol, only three estimates of the radiative forcing are available. Van Dorland et al. (1997) has produced a very speculative radiative forcing estimate of approximately -0.03 Wm-2 for ammonium nitrate, while Adams et al. (2001) derived a radiative forcing of -0.22 Wm-2 from an anthropogenic burden of 0.62 mg(NO3)m-2, and Jacobson (2001) derived a radiative forcing of approximately -0.02 Wm-2 for a column burden of 0.7 mg(NO3)m-2. It appears that the large discrepancy between the results of Jacobson (2001) and Adams et al. (2001) is that 90% of the nitrate is in the coarse mode in Jacobson (2001), which reduces the scattering efficiency. Recent measurement studies by Veefkind et al. (1996) and ten Brink et al. (1997) in the Netherlands have shown that nitrate aerosol in the form of ammonium nitrate is a locally important aerosol species in terms of aerosol mass in the optically active sub-micron size range and hence the associated radiative forcing. They also emphasise the problems in measuring the concentrations and size distributions of nitrate which is a semi-volatile substance. The contradictory nature of the global studies means that no "best estimate" or range for the radiative forcing due to anthropogenic nitrate aerosol is presented in this report, though future studies may prove that it exerts a significant radiative forcing.
While the radiative forcing due to greenhouse gases may be determined to a reasonably high degree of accuracy (Section 6.3), the uncertainties relating to aerosol radiative forcings remain large, and rely to a large extent on the estimates from global modelling studies that are difficult to verify at the present time. The range of estimates presented in this section represents mainly the structural uncertainties (i.e., differences in the model structures and assumptions) rather than the parametric uncertainties (i.e., the uncertainties in the key parameters) (see Pan et al., 1997). This is because many of the model calculations for sulphate aerosol use identical size distributions and refractive indices which lead to identical optical parameters. Thus the model results are not necessarily independent, and certainly do not include the full range of parametric uncertainties. The response of the direct radiative forcing to parametric uncertainties is investigated in sensitivity studies for different aerosol species (e.g., Boucher and Anderson, 1995; Haywood and Ramaswamy, 1998). Three major areas of uncertainty exist; uncertainties in the atmospheric burden and the anthropogenic contribution to it, uncertainties in the optical parameters, and uncertainties in implementation of the optical parameters and burden to give a radiative forcing. The atmospheric burden of an anthropogenic aerosol species is determined by factors such as emission, aging, convective transport, scavenging and deposition processes, each of which have an associated uncertainty (see Chapter 5). The optical parameters have uncertainties associated with uncertainties in size distribution, chemical composition, state of mixing, method of mixing, and asphericity. The problem of the degree of external/internal mixing in the atmosphere deserves highlighting, as global modelling studies tend to assume external mixtures which make modelling the sources, atmospheric transport and radiative properties simpler (e.g., Tegen et al., 1997). However, single particle analysis of particles containing mineral dust (e.g., Levin et al., 1996) and sea salt (Murphy et al., 1998) have often shown them to be internally mixed with sulphate and other aerosols of anthropogenic origin. Thus, heterogeneous conversion of SO2 to sulphate aerosol on dust or sea salt particles may effectively lead to sulphate becoming internally mixed with larger super-micron particles (e.g., Dentener et al., 1996) leading to a reduction in extinction efficiency, an effect that has not yet been accounted for in global modelling studies. Studies that model internal mixtures of absorbing and scattering aerosols necessarily apply simplifying assumptions such as volume weighting the refractive indices (e.g., Haywood et al., 1997a; Myhre et al., 1998c) which may overestimate the degree of absorption (e.g., Jacobson, 2000). Modelling studies that examine the effects of internal mixing of multi-component aerosols are only just becoming available (Jacobson, 2001). Uncertainties in calculating the radiative forcing from specified burdens and optical parameters arise from uncertainties in the parametrization of relative humidity effects, the horizontal and vertical distributions of the aerosol, the uncertainties and sub-grid scale effects in other model fields such as clouds, humidity, temperature and surface reflectance, the representation of the diurnal cycle, and the accuracy of the radiation code used in the calculations. The short atmospheric lifetime of aerosols and the resultant large spatial variability imply a strong requirement for global observational data. Until a reliable global observational method for verifying the radiative effects of anthropogenic aerosols is available, it is likely that the radiative forcing of any aerosol species will remain difficult to quantify. Although satellite retrievals of aerosol optical properties have advanced substantially since the SAR, the difficult problem of separating anthropogenic from natural aerosol still remains (Chapter 5). Nevertheless, new analyses reiterate that global satellite measurements contain tropospheric aerosol signatures that include those due to anthropogenic aerosols (Haywood et al., 1999; Boucher and Tanré, 2000). While the general spatial distribution of the radiative forcing for sulphate aerosol appears to be similar among the studies listed in Table 6.4, some important features, such as the seasonal cycle in the radiative forcing, remain highly uncertain which may have important consequences in terms of the detection and attribution of climate change (Chapter 12).
Here we examine the consistency of the ranges derived in this section with the ranges obtained using the approach of Chapter 5, Section 5.4.2. For the industrial sulphate, fossil fuel BC, and fossil fuel OC aerosols, the "best estimates" of the direct radiative forcing are -0.4 Wm-2 with a factor of 2 uncertainty, +0.2 Wm-2 with a factor of 2 uncertainty and -0.1 Wm-2 with a factor of 3 uncertainty, respectively. An estimate of the total direct radiative forcing from industrial aerosols (using RMS errors) leads to a range -0.07 to -1.24 Wm-2 which is reasonably consistent with -0.1 to -1.0 Wm-2 (one standard deviation) from Chapter 5, Section 5.4.2. For biomass burning aerosols the "best estimate" of the direct radiative forcing is -0.2 Wm-2 with a range -0.07 to -0.6 Wm-2, which is reasonably consistent with -0.1 to -0.5 Wm-2 (one standard deviation) obtained from Section 5.4.2.
Additionally, while this section has concentrated upon the radiative forcing at the top of the atmosphere, the effects of anthropogenic aerosols upon the radiative budget at the surface of the Earth has not been considered in detail. For purely scattering aerosols in cloud-free conditions, the radiative effect at the surface is within a few per cent of that at the top of the atmosphere (e.g., Haywood and Shine, 1997). However, for partially absorbing aerosols, the radiative effect at the surface may be many times that at the top of the atmosphere, as evidenced by modelling and measurement studies (e.g., Haywood and Shine, 1997; Haywood et al., 1999; Chapter 5). This is because, for partially absorbing aerosols, energy is transferred directly to the atmospheric column. Ackerman et al. (2000) point out that this can warm the atmosphere and "burn off" clouds. They conclude that during the northeast monsoon (dry season over India) daytime trade cumulus cloud cover over the northern Indian Ocean can be reduced by nearly half, although these results depend strongly upon the meteorological conditions and modelling assumptions. This process may also be important over the global domain as indicated by Hansen et al. (1997a), as the climate sensitivity parameter (Section 6.2) may differ significantly for absorbing aerosols due to diabatic heating in the aerosol layer modifying the temperature structure of the atmosphere, which affects the formation of clouds. The vertical partitioning of the forcing by absorbing aerosols is also a potentially important factor in determining climatic changes at the surface (e.g., evaporation, soil moisture).
Other reports in this collection