Tropospheric black carbon (BC) aerosol, also described as soot, primarily absorbs incident solar radiation, which leads to positive radiative forcing. As in the case of sulfate aerosol, the small size of the particles means that radiative forcing in the longwave region of the spectrum is likely to be negligible. The sensitivity of global mean radiative forcing to column loading of total anthropogenic BC is estimated by Haywood et al. (1997a), Haywood and Ramaswamy (1998), and Myhre et al. (1998) to range from approximately +1100 to +1850 W g-1 BC.
We reexamined results by inserting BC aerosol in a layer between 8 and 13 km in the GFDL R30 GCM using the method of Haywood and Ramaswamy (1998). A log-normal distribution with a geometric mean radius of 0.0118 µm and a standard deviation of 2.0 was assumed. The resulting global mean sensitivity was found to be approximately +3000 W g-1 BC as a result of the higher sensitivity of the radiative forcing when the BC exists at higher altitudes above a greater proportion of cloudy layers (Haywood and Ramaswamy, 1998). This value is adopted throughout this report because it explicitly takes into account the effect of the elevated altitude of the aerosol.
BC particles primarily absorb sunlight and heat the local air. Thus, unlike BC that resides in the troposphere, BC in the stratosphere contributes negative solar radiative forcing that is countered by induced positive longwave radiative forcing. Thus, radiative forcing from BC aerosols is sensitive to their location relative to the tropopause. We do not have enough information on the location of BC relative to the tropopause, and thus our use of the instantaneous top-of-atmosphere value overestimates the RF depending on the fraction of aircraft BC in the lower stratosphere.
Using the aircraft fuel-burn scenarios for NASA-1992 noted above and described in Chapter 3, we derive a global mean column burden of BC aerosol from aircraft of 1.0 µg BC m-2 (assuming an EI(BC) of 0.04 g kg-1; see also Table 3-4). Thus, we estimate global mean BC aerosol forcing in 1992 to be +0.003 (+0.001 to +0.006) W m-2 and assume that it linearly scales with fuel use (see also Table 6-1). This value is much smaller in absolute magnitude than the RF from CO2, O3, CH4, or contrails.
Aircraft emission of water vapor and particles, as well as the creation of contrails, could lead to a change in global cloudiness. Some atmospheric GCM studies that have looked at the impacts of injecting water vapor or creating contrails (e.g., Ponater et al., 1996; Rind et al. 1996) point to the potential importance of these effects on climate, but these pilot studies cannot be used directly in this assessment. Persistent contrails clearly related to aircraft are detectable, however, and their impact on radiative forcing can be evaluated. Section 3.6 (see Table 3-9) estimates direct radiative forcing from persistent contrails to be +0.02 (+0.005 to +0.06) W m-2 in 1992 (see also Table 6-1). This estimate is limited to immediately visible, quasi-linear persistent contrails.
Whereas contrail formation and associated radiative forcing is an obvious and visible consequence of aircraft activity, the secondary, indirect effect of aerosols from aircraft on the microphysical and radiative properties of clouds is a very complex issue that has received little attention and is very difficult to quantify (Seinfeld, 1998). Some significant steps in quantifying the indirect effect from anthropogenic aerosols have been made (e.g., Jones et al., 1994; Boucher and Lohmann, 1995). The effects of aerosol particles from aircraft emissions on clouds are more complicated because nucleation and subsequent growth of ice crystals that make up cirrus clouds are more complex and less studied than for water clouds. Cirrus cloud generally exert positive forcing because longwave positive radiative forcing is of a larger magnitude than solar negative radiative forcing. Section 3.6.5 (see Table 3-9) estimates that radiative forcing from aircraft-induced cirrus is positive and may be comparable to contrail RF. The magnitude of this RF remains very uncertain. No best estimate is given in Tables 6-1 and 6-2, but a range for the best estimate could fall between 0 and 0.04 W m-2.
Other reports in this collection