IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis Black Carbon Aerosol from Fossil Fuels

Black carbon (BC) is a primary aerosol emitted directly at the source from incomplete combustion processes such as fossil fuel and biomass burning and therefore much atmospheric BC is of anthropogenic origin. Global, present-day fossil fuel emission estimates range from 5.8 to 8.0 TgC yr–1 (Haywood and Boucher, 2000 and references therein). Bond et al. (2004) estimated the total current global emission of BC to be approximately 8 TgC yr–1, with contributions of 4.6 TgC yr–1 from fossil fuel and biofuel combustion and 3.3 TgC yr–1 from open biomass burning, and estimated an uncertainty of about a factor of two. Ito and Penner (2005) suggested fossil fuel BC emissions for 2000 of around 2.8 TgC yr–1. The trends in emission of fossil fuel BC have been investigated in industrial areas by Novakov et al. (2003) and Ito and Penner (2005). Novakov et al. (2003) reported that significant decreases were recorded in the UK, Germany, the former Soviet Union and the USA over the period 1950 to 2000, while significant increases were reported in India and China. Globally, Novakov et al. (2003) suggested that emissions of fossil fuel BC increased by a factor of three between 1950 and 1990 (2.2 to 6.7 TgC yr–1) owing to the rapid expansion of the USA, European and Asian economies (e.g., Streets et al., 2001, 2003), and have since fallen to around 5.6 TgC yr–1 owing to further emission controls. Ito and Penner (2005) determined a similar trend in emissions over the period 1950 to 2000 of approximately a factor of three, but the absolute emissions are smaller than in Novakov et al. (2003) by approximately a factor of 1.7.

Black carbon aerosol strongly absorbs solar radiation. Electron microscope images of BC particles show that they are emitted as complex chain structures (e.g., Posfai et al., 2003), which tend to collapse as the particles age, thereby modifying the optical properties (e.g., Abel et al., 2003). The Indian Ocean Experiment (Ramanathan et al., 2001b and references therein) focussed on emissions of aerosol from the Indian sub-continent, and showed the importance of absorption by aerosol in the atmospheric column. These observations showed that the local surface forcing (–23 W m–2) was significantly stronger than the local RF at the TOA (–7 W m–2). Additionally, the presence of BC in the atmosphere above highly reflective surfaces such as snow and ice, or clouds, may cause a significant positive RF (Ramaswamy et al., 2001). The vertical profile is therefore important, as BC aerosols or mixtures of aerosols containing a relatively large fraction of BC will exert a positive RF when located above clouds. Both microphysical (e.g., hydrophilic-to-hydrophobic nature of emissions into the atmosphere, aging of the aerosols, wet deposition) and meteorological aspects govern the horizontal and vertical distribution patterns of BC aerosols, and the residence time of these aerosols is thus sensitive to these factors (Cooke et al., 2002; Stier et al., 2006b).

The TAR assessed the RF due to fossil fuel BC as being +0.2 W m–2 with an uncertainty of a factor of two. Those models since the TAR that explicitly model and separate out the RF due to BC from fossil fuels include those from Takemura et al. (2000), Reddy et al. (2005a) and Hansen et al. (2005) as summarised in Table 2.5. The results from a number of studies that continue to group the RF from fossil fuel with that from biomass burning are also shown. Recently published results (A to K) and AeroCom studies (L to T) suggest a combined RF from both sources of +0.44 ± 0.13 W m–2 and +0.29 ± 0.15 W m–2 respectively. The stronger RF estimates from the models A to K appear to be primarily due to stronger sources and column loadings as the direct RF/column loading is similar at approximately 1.2 to 1.3 W mg–1 (Table 2.5). Carbonaceous aerosol emission inventories suggest that approximately 34 to 38% of emissions come from biomass burning sources and the remainder from fossil fuel burning sources. Models that separate fossil fuel from biomass burning suggest an equal split in RF. This is applied to those estimates where the BC emissions are not explicitly separated into emission sources to provide an estimate of the RF due to fossil fuel BC. For the AeroCom results, the fossil fuel BC RF ranges from +0.08 to +0.18 W m–2 with a mean of +0.13 W m–2 and a standard deviation of 0.03 W m–2. For model results A to K, the RFs range from +0.15 W m–2 to approximately +0.27 W m–2, with a mean of +0.25 W m–2 and a standard deviation of 0.11 W m–2.

The mean and median of the direct RF for fossil fuel BC from grouping all these studies together are +0.19 and +0.16 W m–2, respectively, with a standard deviation of nearly 0.10 W m–2. The standard deviation is multiplied by 1.645 to approximate the 90% confidence interval and the best estimate is rounded upwards slightly for simplicity, leading to a direct RF estimate of +0.20 ± 0.15 W m–2. This estimate does not include the semi-direct effect or the BC impact on snow and ice surface albedo (see Sections 2.5.4 and