Open Access

Dominick V. Spracklen, Kenneth S. Carslaw, Joonas Merikanto, Graham W. Mann, Steven Pickering, John A. Ogren, Elisabeth Andrews, Urs Baltensperger, Ernest Weingartner, Michael Boy, Markku Kulmala, Lauri Laakso, Heikki Lihavainen, Niku Kivekäs, Nikos Mihalopoulos, G. Kouvarakis, S. Gerard Jennings, Wolfram Birmili, Alfred Wiedensohler, Rolf Weller, Paolo Laj, Karine Sellegri, Boris Bonn, Radek Krejci

• We use observations of total particle number concentration at 36 worldwide sites and a global aerosol model to quantify the primary and secondary sources of particle number. We show that emissions of primary particles can reasonably reproduce the spatial pattern of observed condensation nuclei (CN) (R2=0.51) but fail to explain the observed seasonal cycle at many sites (R2=0.1). The modeled CN concentration in the free troposphere is biased low (normalised mean bias, NMB=−88%) unless a secondary source of particles is included, for example from binary homogeneous nucleation of sulfuric acid and water (NMB=−25%). Simulated CN concentrations in the continental boundary layer (BL) are also biased low (NMB=−74%) unless the number emission of anthropogenic primary particles is increased or an empirical BL particle formation mechanism based on sulfuric acid is used. We find that the seasonal CN cycle observed at continental BL sites is better simulated by including a BL particle formation mechanism (R2=0.3) than by increasing the number emission from primary anthropogenic sources (R2=0.18). Using sensitivity tests we derive optimum rate coefficients for this nucleation mechanism, which agree with values derived from detailed case studies at individual sites.