2.4.4.2 Organic Carbon Aerosol from Fossil Fuels

Organic aerosols are a complex mixture of chemical compounds containing carbon-carbon bonds produced from fossil fuel and biofuel burning and natural biogenic emissions. Organic aerosols are emitted as primary aerosol particles or formed as secondary aerosol particles from condensation of organic gases considered semi-volatile or having low volatility. Hundreds of different atmospheric organic compounds have been detected in the atmosphere (e.g., Hamilton et al., 2004; Murphy, 2005), which makes definitive modelling of the direct and indirect effects extremely challenging (McFiggans et al., 2006). Emissions of primary organic carbon from fossil fuel burning have been estimated to be 10 to 30 TgC yr^–1 (Liousse et al., 1996; Cooke et al., 1999; Scholes and Andreae, 2000). More recently, Bond et al. (2004) provided a detailed analysis of primary organic carbon emissions from fossil fuels, biofuels and open burning, and suggested that contained burning (approximately the sum of fossil fuel and biofuel) emissions are in the range of 5 to 17 TgC yr^–1, with fossil fuel contributing only 2.4 TgC yr^–1. Ito and Penner (2005) estimated global fossil fuel particulate organic matter (POM, which is the sum of the organic carbon and the other associated chemical elements) emissions of around 2.2 Tg(POM) yr^–1, and global biofuel emissions of around 7.5 Tg(POM) yr^–1. Ito and Penner (2005) estimated that emissions of fossil and biofuel organic carbon increased by a factor of three over the period 1870 to 2000. Subsequent to emission, the hygroscopic, chemical and optical properties of organic carbon particles continue to change because of chemical processing by gas-phase oxidants such as ozone, OH, and the nitrate radical (NO₃) (e.g., Kanakidou et al., 2005). Atmospheric concentrations of organic aerosol are frequently similar to those of industrial sulphate aerosol. Novakov et al. (1997) and Hegg et al. (1997) measured organic carbon in pollution off the East Coast of the USA during the TARFOX campaign, and found organic carbon primarily from fossil fuel burning contributed up to 40% of the total submicron aerosol mass and was frequently the most significant contributor to τ_aer. During INDOEX, which studied the industrial plume over the Indian Ocean, Ramanathan et al. (2001b) found that organic carbon was the second largest contributor to τ_aer after sulphate aerosol.

Observational evidence suggests that some organic aerosol compounds from fossil fuels are relatively weakly absorbing but do absorb solar radiation at some ultraviolet and visible wavelengths (e.g., Bond et al., 1999; Jacobson, 1999; Bond, 2001) although organic aerosol from high-temperature combustion such as fossil fuel burning (Dubovik et al., 1998; Kirchstetter et al., 2004) appears less absorbing than from low-temperature combustion such as open biomass burning. Observations suggest that a considerable fraction of organic carbon is soluble to some degree, while at low relative humidity more water is often associated with the organic fraction than with inorganic material. At higher relative humidities, the hygroscopicity of organic carbon is considerably less than that of sulphate aerosol (Kotchenruther and Hobbs, 1998; Kotchenruther et al., 1999).

Based on observations and fundamental chemical kinetic principles, attempts have been made to formulate organic carbon composition by functional group analysis in some main classes of organic chemical species (e.g., Decesari et al., 2000, 2001; Maria et al., 2002; Ming and Russell, 2002), capturing some general characteristics in terms of refractive indices, hygroscopicity and cloud activation properties. This facilitates improved parametrizations in global models (e.g., Fuzzi et al., 2001; Kanakidou et al., 2005; Ming et al., 2005a).

Organic carbon aerosol from fossil fuel sources is invariably internally and externally mixed to some degree with other combustion products such as sulphate and black carbon (e.g., Novakov et al., 1997; Ramanathan et al., 2001b). Theoretically, coatings of essentially non-absorbing components such as organic carbon or sulphate on strongly absorbing core components such as black carbon can increase the absorption of the composite aerosol (e.g., Fuller et al., 1999; Jacobson, 2001a; Stier et al., 2006a), with results backed up by laboratory studies (e.g., Schnaiter et al., 2003). However, coatings of organic carbon aerosol on hygroscopic aerosol such as sulphate may lead to suppression of the rate of water uptake during cloud activation (Xiong et al., 1998; Chuang, 2003).

Current global models generally treat organic carbon using one or two tracers (e.g., water-insoluble tracer, water-soluble tracer) and highly parametrized schemes have been developed to represent the direct RF. Secondary organic carbon is highly simplified in the global models and in many cases treated as an additional source similar to primary organic carbon. Considerable uncertainties still exist in representing the refractive indices and the water of hydration associated with the particles because the aerosol properties will invariably differ depending on the combustion process, chemical processing in the atmosphere, mixing with the ambient aerosol, etc. (e.g., McFiggans et al., 2006).

The TAR reported an RF of organic carbon aerosols from fossil fuel burning of –0.10 W m^–2 with a factor of three uncertainty. Many of the modelling studies performed since the TAR have investigated the RF of organic carbon aerosols from both fossil fuel and biomass burning aerosols, and the combined RF of both components. These studies are summarised in Table 2.5. The RF from total organic carbon (POM) from both biomass burning and fossil fuel emissions from recently published models A to K and AeroCom models (L to T) is –0.24 W m^–2 with a standard deviation of 0.08 W m^–2 and –0.16 W m^–2 with a standard deviation of 0.10 W m^–2, respectively. Where the RF due to organic carbon from fossil fuels is not explicitly accounted for in the studies, an approximate scaling based on the source apportionment of 0.25:0.75 is applied for fossil fuel organic carbon:biomass burning organic carbon (Bond et al., 2004). The mean RF of the fossil fuel component of organic carbon from those studies other than in AeroCom is –0.06 W m^–2, while those from AeroCom produce an RF of –0.03 W m^–2 with a range of –0.01 W m^–2 to –0.06 W m^–2 and a standard deviation of around 0.02 W m^–2. Note that these RF estimates, to a large degree, only take into account primary emitted organic carbon. These studies all use optical properties for organic carbon that are either entirely scattering or only weakly absorbing and hence the surface forcing is only slightly stronger than that at the TOA.

The mean and median for the direct RF of fossil fuel organic carbon from grouping all these studies together are identical at –0.05 W m^–2 with a standard deviation of 0.03 W m^–2. The standard deviation is multiplied by 1.645 to approximate the 90% confidence interval.^[9] This leads to a direct RF estimate of –0.05 ± 0.05 W m^–2.