Abstract: In order to understand the characteristics of particulate pollution in recent sand dust weather in Tianjin, characterization of particulate matter pollution during four typical sand dust weather processes (nine dust days ) in 2018 were investigated based on high-resolution online monitoring data and backward track model. The results show that the hourly peak concentrations of TSP and PM₁₀ were 2 265 and 928 μg·m^-3, and the average concentrations were 970 and 435 μg·m^-3, respectively. The mass ratios of PM₁₀ to TSP were 0.45-0.83, and the mass ratios of PM_2.5 to PM₁₀ were 0.17-0.29. The number concentrations of particulate matter decreased rapidly, and the distribution trend of particulate spectrum show that the peak of particulate number concentrations decreased, and shifted to the direction of coarse particles. The peak size spectrum of the dust was among 0.67-1.72 μm. Chemical compositions concentrations of PM_2.5 showed different change characteristics during sand dust weather processes. Concentrations of cations such as Ca²⁺, Na⁺ and Mg²⁺ increased significantly, while the concentrations of secondary ions (NO₃^-、SO₄^2-、NH₄⁺) decreased significantly. The extinction coefficients were smaller and the depolarization ratios were larger at altitude below 5 km of sand dust weather, and the depolarization ratios increased obviously with time and gradually transferred to the ground. As for four typical sand dust weather processes in 2018 in Tianjin, the sand dust mass was transported to Beijing-Tianjin-Hebei Region from an altitude of about 3 km with the northern path or the northwest path influence of Tianjin, and its main sources located at central, south and southeast part of Mongolia.

Key words: sand dust weather, particulate matter, size spectrum, water soluble ions, Lidar, backward trace model