Quantitatively analyzing effects of meteorology and PM_2.5 sources on low visual distance

Low visual distance (0 to 10 km), a common pollution phenomenon, is a severe threat to the productivity and life of human society in China. In this study, we used the Positive Matrix Factorization model coupled with the Generalized Additive Model (PMF-GAM) to quantitatively analyze the effect of the meteorology and source emissions on visual distance. The results show that the relative importance of predictor variables is humidity and SN (sulfate and secondary organic carbon (SOC) plus nitrate) (H&SN, 69.14%), vehicle exhaust (VE, 13.5%), crustal dust (CD, 7.28%), temperature (T, 4.71%), coal combustion (CC, 4.08%), wind speed (WS, 1.08%) and atmospheric pressure (AP, 0.21%). Furthermore, the visual distance is higher when the humidity is lower (<20%), and the humidity with SN shows clear synergy effects when the humidity is higher (>60%).