Measurements of submicron particles vertical profiles by means of topographic relief in a typical valley city, China.

Abstract To reveal PM 1 vertical profiles and key affecting factors in a typical urban valley, daytime and nighttime PM 1 (the particles with diameters less than 1 μm) samples were collected with medium volume air samplers during 26 December 2017 to 11 January 2018 at five different altitudes by means of high topographic relief at urban areas of Lanzhou. The synchronous boundary layer temperature and humidity profiles were observed by a microwave radiometer. Daytime PM 1 concentrations reduced by about 3.86 μg m−3 when the height above the surface increased by 100 m, which was much lower than that for nighttime (5.68 μg m−3 100 m−1) as particles were easily accumulated near the surface when the air was stable during the nighttime. The three typical PM 1 vertical profiles were identified by K-means clustering technique. The most frequent cluster with elevated PM 1 concentrations near the surface was closely related to temperature inversion around the ground, while the cluster with relatively uniform PM 1 within the boundary layer was mainly induced by unstable atmospheric stratification and thus relatively good vertical dispersion. About 50%–60% of PM 1 variations could be attributed to atmospheric stratification near the surface in the valley city, which was much higher than that at the hilltop. The PM 1 difference increased by 47.14 (36.91) μg m−3 when inversion layer thickness (intensity) increased by 100 m (1 °C 100 m−1). The newly calculated inversion index considering both inversion layer thickness and intensity explained about 87% of PM 1 differences between near the surface and at the hilltop. The vertical dispersion had a more significant effect on PM 1 than horizontal dispersion near the surface, while PM 1 was more largely affected by horizontal dispersion at the hilltop, which was closely related to the valley terrain. Highlights • PM 1 vertical profiles were revealed in urban areas of Lanzhou for the first time. • Three typical PM 1 profile types were identified using K-means clustering technique. • PM 1 concentrations reduced by about 3–5 μg m−3 when the height increased by 100 m. • 50%–60% of PM 1 variations were attributed to atmospheric stratification near the surface. • PM 1 was more largely affected by vertical dispersion near the surface. [ABSTRACT FROM AUTHOR]