Seasonal and regional variations of source contributions for PM10 and PM2.5 in urban environment.

Abstract To characterize the sources of to PM 10 and PM 2.5 , a long-term, speciate and simultaneous dataset was sampled in a megacity in China during the period of 2006–2014. The PM concentrations and PM 2.5 /PM 10 were higher in the winter. Higher percentages of Al, Si, Ca and Fe were observed in the summer, and higher concentrations of OC, NO 3 − and SO 4 2 − occurred in the winter. Then, the sources were quantified by an advanced three-way model (defined as an ABB three-way model), which estimates different profiles for different sizes. A higher percentage of cement and crustal dust was present in the summer; higher fractions of coal combustion and nitrate + SOC were observed in the winter. Crustal and cement contributed larger portion to coarse part of PM 10 , whereas vehicular and secondary source categories were enriched in PM 2.5 . Finally, potential source contribution function (PSCF) and source regional apportionment (SRA) methods were combined with the three-way model to estimate geographical origins. During the sampling period, the southeast region (R4) was an important region for most source categories (0.6%–11.5%); the R1 (centre region) also played a vital role (0.3–6.9%).