Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

6.7.5 Biomass Burning Aerosol

Biomass burning aerosol consists of two major chemical components: black carbon (BC), which primarily absorbs solar radiation, and organic carbon (OC), which primarily scatters solar radiation. Sources of biomass burning aerosol include burning of forests and savanna for colonisation and agriculture, burning of agricultural waste, and substances burned for fuel such as wood, dung and peat. Not all biomass aerosol comes from anthropogenic activities, as naturally occurring vegetation fires regularly occur. The fraction of anthropogenic biomass aerosol remains difficult to deduce. As for sulphate and fossil fuel BC aerosol, three-dimensional GCM estimates (e.g., Penner et al., 1998b) have superseded earlier simple box-model calculations (e.g., Penner et al., 1992) (see Table 6.5). Penner et al. (1998b) and Grant et al. (1999) used a GCM to model the radiative forcing from biomass aerosol, finding a combined total radiative forcing of -0.14 to -0.23 Wm-2 depending upon the mixing assumptions and size distributions used. The radiative forcing is calculated to be negative over the majority of the globe, but some limited areas of positive forcing exist over areas with a high surface reflectance (see Section 6.14.2). The seasonal cycle is strongly influenced by the seasonal cycle of biomass burning emissions (Grant et al., 1999; Iacobellis et al., 1999), the global mean being estimated to be a maximum during June/July/August. Hobbs et al. (1997) performed aircraft measurements of smoke from biomass burning during the Smoke Cloud and Radiation-Brazil (SCAR-B) experiment and used the model of Penner et al. (1992) to estimate a global mean radiative forcing of -0.3 Wm-2 as an approximate upper limit due to modified optical parameters. Ross et al. (1998) performed a similar measurement and modelling study estimating a local annual mean radiative forcing of -2 to -3 Wm-2 in intensive biomass burning regions, indicating that the global mean radiative forcing is likely to be significantly smaller than in Penner et al. (1992). Iacobellis et al. (1999) model the global radiative forcing as -0.7 Wm-2 but use an emission factor for biomass aerosols that Liousse et al. (1996) suggest is a factor of three too high, thus a radiative forcing of -0.25 Wm-2 is more likely. These estimates neglect any long-wave radiative forcing although Christopher et al. (1996) found a discernible signal in AVHRR data from Brazilian forest fires that opposes the short-wave radiative forcing; the magnitude of the long-wave signal will depend upon the size, optical parameters, and altitude of the aerosol.

The radiative effects of the individual BC and OC components from biomass burning have also received attention. Haywood and Ramaswamy (1998) re-scaled the global column burden of BC from Cooke and Wilson (1996) to that of Liousse et al. (1996) which is thought to be more representative of optically active BC and estimated the radiative forcing due to fossil fuel and biomass burning to be approximately +0.4 Wm-2, approximately half of which, or +0.2 Wm-2, is due to biomass sources. Hansen et al. (1998) adjusted the single scattering albedo of background aerosols from unity to 0.92 to 0.95 to account for the absorbing properties of BC from combined fossil fuel and biomass emissions and found a radiative forcing of +0.27 Wm-2.

The radiative forcing due to the purely scattering OC component from combined fossil fuel and biomass emissions is estimated by Hansen et al. (1998) to be -0.41 Wm-2 with a approximate upper limit for the radiative forcing due to fossil fuel OC from this study of approximately -0.16 Wm-2 (see Section 6.7.4). Thus an approximate weakest limit for the radiative forcing due to biomass burning OC is -0.25 Wm-2. Jacobson (2001) used a multi-component aerosol model to investigate the radiative forcing due to OC from combined fossil fuel and biomass aerosols using emissions from Liousse et al. (1996) finding a resultant radiative forcing of between -0.04 and -0.06 Wm-2, the strongest forcing being for purely scattering aerosol. The small radiative forcings of Jacobson (2001) may be due to the aerosol being modelled as tri-modal with less aerosol in the optically active region of the spectrum, or due to the fact that absorption by BC may be enhanced in the modelled multi-component mixture.

On the basis of these studies, the estimate of the radiative forcing due to biomass burning aerosols remains the same at -0.2 Wm-2. The uncertainty associated with the radiative forcing is very difficult to estimate due to the limited number of studies available and is estimated as at least a factor of three, leading to a range of -0.07 to -0.6 Wm-2.



Other reports in this collection