Marine aerosol production: a review of the current knowledge

0 TNO , PO Box 96864, 2509 JG, The Hague , The Netherlands 1 Climate and Global Change Unit, Research and Development, Finnish Meteorological Institute , 00560 Helsinki , Finland 2 Department of Experimental Physics and Environmental Change Institute, National University of Ireland , Galway University Road, Galway , Ireland The current knowledge in primary and secondary marine aerosol formation is reviewed. For primary marine aerosol source functions, recent source functions have demonstrated a significant flux of submicrometre particles down to radii of 20 nm. Moreover, the source functions derived from different techniques up to 10 mm have come within a factor of two of each other. For secondary marine aerosol formation, recent advances have identified iodine oxides and isoprene oxidation products, in addition to sulphuric acid, as contributing to formation and growth, although the exact roles remains to be determined. While a multistep process seems to be required, isoprene oxidation products are more likely to participate in growth and sulphuric acid is more likely to participate in nucleation. Iodine oxides are likely to participate in both nucleation and growth. 1. Introduction The marine aerosol constitutes one of the most important natural aerosol systems globally. It contributes significantly to the Earths radiative budget, biogeochemical cycling, impacts on ecosystems and even to regional air quality. The marine aerosol comprises primary and secondary aerosol components. The primary aerosol production results from the interaction of wind stress at the ocean surface and results in the mechanical production of sea-spray aerosol (sea spray being the combination of inorganic sea salt and organic matter). Sea spray is produced via the bubble-bursting process typically resulting from whitecap generation, producing film and jet drops, resulting in sea-spray particles in the range of the submicrometre size and up to a few micrometres. It is estimated that whitecap formation onset occurs at wind speeds of 4 m sK1. At higher wind speeds, the direct tearing of wave crests can result if spume droplet formation occurs at sizes from tens to hundreds of micrometres. Globally, in terms of sea-salt production, the mass flux is estimated to be 13!1016 g yrK1 (Erickson & Duce 1988; Gong et al. 2002). Thirty-two per cent of the global flux comes from the Northern Hemisphere and 92% of the mass flux is attributed to the supermicrometre size range (i.e. rO0.5 mm). Increase in sea-salt mass is primarily associated with increasing wind speeds with mass concentrations being measured up to 1000 mg mK3 at wind speeds of 1520 m sK1 (Lewis & Schwartz 2004). Although mass loadings can be high, marine aerosol number concentrations are typically low, of the order of 300600 cmK3 (ODowd et al. 1997; Yoon et al. 2007), of which up to 50150 cmK3 can be sea-salt particles under high wind conditions (ODowd & Smith 1993; Kreidenweis et al. 1998). The interest in sea-spray aerosol has traditionally been focused on large particles (O1 mm) owing to their influence, at high wind speeds, on seaair transfer of water vapour and latent heat (e.g. Andreas 1998). The overview of sea-spray source fluxes presented by Andreas (2002), which includes most of the common source functions presented until 1998, shows a variation of approximately six orders of magnitude. A critical analysis of the source functions presented in Andreas (2002) is discussed in Schulz et al. (2004). More recently, the interest in sea spray has been shifted to other applications such as its role in chemical reactions (for coupled nitric acidsea salt see Sorensen et al. (2005); for coupled sulphatesea salt cycles see ODowd et al. (1999a, 2000)) and, in particular, its role in climate change (IPCC 2001). Sea salt is the dominant submicrometre scatterer in most ocean regions (e.g. Kleefeld et al. 2002; Bates et al. 2006) and dominates the marine boundary layer (MBL) particulate mass concentration in remote oceanic regions, with a significant fraction occurring in the submicrometre size range (IPCC 2001). Sea salt contributes 44% to the global aerosol optical depth. Estimates for topof-atmosphere, global-annual radiative forcing due to sea salt are K1.51 and K5.03 W mK2 for low and high emission values, respectively (IPCC 2001). Sea spray not only affects climate by scattering of solar radiation, but also the spray particles act as cloud condensation nuclei and thus contribute to the indirect aerosol effect (ODowd et al. 1999b). Sea salt has also been linked to the MBL cycle through the activation of halogens, leading to ozone depletion ( Vogt et al. 1996; McFiggens et al. 2000). For primary marine aerosol (PMA), a historical and detailed review of sea-salt production and resulting concentrations is discussed in Lewis & Schwartz (2005). In terms of secondary marine aerosol production (i.e. particle production resulting from gas-to-particle conversion processes), historically, it has been thought that sulphur species have been the primary chemical component involved (Shaw 1983; Charlson et al. 1987). Secondary aerosol production occurs in the following two ways: (i) new particle formation occurs via the nucleation of stable clusters of the order of 0.51 nm in sizes (these clusters, once formed, can grow to larger sizes via condensation processes) and (ii) they also can grow via heterogeneous reactions and aqueous phase oxidation of dissolved gases in existing aerosol particles. In terms of the sulphur cycle, dimethylsulphide (DMS), a waste produced by phytoplankton, is released from the ocean into the atmosphere where it undergoes oxidation by the OH radical to form SO2, which is further oxidized to form H2SO4 (see Charlson et al. (1987) for full discussion). H2SO4 is thought to participate in binary homogeneous nucleation with H2O, and more recently, in ternary nucleation with H2O and NH3. While Clarke et al. (1998) found evidence for particle production linked to high DMS emissions in the Pacific, a more robust analysis of particle production from H2SO4 by Pirjola et al. (2000) pointed out that while binary nucleation was likely to occur in the polar regions, and ternary nucleation likely to occur in many other marine environments, there is typically insufficient H2SO4 vapour to contribute to growth of stable clusters into aerosol particles (operationally defined as particles with DO3 nm). ODowd et al. (2002a) demonstrated that unless newly formed stable clusters can grow sufficiently fast, the clusters are scavenged by the preexisting aerosol coagulation sink. For example, a 1 nm particle, under typical H2SO4 cKo4nncemntsrations, will grow by condensation of H2SO4 vapour at a rate of 0.5!10 K1. This is compared with a coagulation loss rate of 2!10K3 sK1, that is, there is a very low probability of survival for 1 nm particles. If, however, the source rate of condensable vapours is increased significantly above that for H2SO4, and the (...truncated)