Speciated atmospheric mercury including reactive gaseous mercury (RGM), particle bound mercury (PBM), and gaseous elemental mercury (GEM) were measured for a two-year period in Ningbo city, the Yangtze River Delta, China. Averaged concentrations of PBM and RGM were 316 ± 377 pg m^-3 and 100 ± 123 pg m^-3, respectively, with the highest RGM/PBM ratio in summer (0.73 ± 1.31) and the lowest in winter (0.35 ± 0.56). The relationship between RGM/PBM ratio and PM_2.5 was nonlinear. When PM_2.5 fell in the range of 15–100 μg m^-3, the PBM increased, the RGM decreased, and the RGM/PBM ratio showed an obvious decreasing trend as PM_2.5 increased. An empirical linear relationship between log(1/Kp) and 1/T was obtained for different seasons. The correlation coefficient and slope of the equation were highest in winter, suggesting that T explained more for Kp variation and RM transferred quickly to the particle phase in the cold season. The relationships between Kp and aerosol compositions show that Kp increased with increasing fractions of Organics (Org), nitrate (NO₃), and chloride (Chl) and decreasing fractions of sulfate (SO₄) and ammonium (NH₄) in aerosols. Principle component analysis - multiple linear regression (PCA-MLR) was used to evaluate the relative contributions of gas-particle partitioning of RM and sectional primary emissions to PBM in different seasons. The results show that the contributions of RM partitioning to PBM were 28.0% in spring, 28.3% in summer, 31.2% in autumn, and 39.2% in winter, which was well explained by the seasonal variations of temperature and aerosol compositions.

Regression analyses of log(1/Kp) as a function of 1/T (a) during the whole study period and (b) in different seasons.