Abstract. Source apportionment was performed using size segregated particle number concentrations (PNCs) in 27 size channels over the diameter range of 6–1000 nm augmented by air pollutants all with a time resolution of 1 h in the urban background of Budapest for 11 full years in separate seasons. The input dataset was corrected for the effect of the local meteorology by dispersion normalization using the ventilation coefficient defined as the planetary boundary mixing layer height multiplied by the wind speed. Both the uncorrected and dispersion-corrected datasets were evaluated using positive matrix factorization. Six source types including nucleation, two road vehicle emission sources separated into a semi-volatile fraction and a solid core fraction, diffuse urban source, secondary inorganic aerosol (SIA), and ozone-associated particles were identified, characterised, and quantified. The ventilation correction substantially modified the input concentrations, while the differences in the corrected-to-uncorrected ratios for the contributions remained within 5 %. The overall mean relative contribution of the road traffic emission sources was 60 %, and did not show considerable seasonal variability. Nucleation was responsible for 20 % of the PNC annually as a lower estimate. It exhibited a compound character consisting of photochemically induced nucleation and traffic-related nucleation. The former process occurs on regional or urban spatial scales around noon, whereas the latter process happens when the gas-phase vapours in the vehicle exhaust cool, and the resulted supersaturated vapours nucleate outside the source. Its relative contributions were maximal in spring (somewhat smaller in summer and autumn) and minimal in winter. The contributions from the SIA and the urban diffuse source types were approximately 10 % in spring, summer, and 12–15 % in autumn and winter, respectively. The O3-associated secondary aerosol made up the smallest (6 %) portion of particles on an annual basis. Directionality variations investigated by conditional bivariate probability function analysis were used to locate the likely source areas, and showed considerable spatial variations in the source origin.