Re-evaluating the distribution and variation characteristics of haze in China using different distinguishing methods...

Haze is identified via different methods using hourly visibility, relative humidity (RH) and PM_2.5 mass concentration observations collected from 2013 to 2018 at 502 stations in China. An inter-comparison of a new haze identification method (MGB) and other currently used methods (M80 and M90) is performed in this research. Compared with other methods, the MGB method has an advantage in the expression of fine particle pollution characteristics, especially in high humidity areas. The mean value of the correlation coefficient of the daily mean PM_2.5 and daily haze hour obtained by MGB in China is 0.69 which is higher than the correlation coefficients of the daily mean PM_2.5 and haze hour identified by the other two methods. Compared with M80, the haze identified by MGB and M90 is less influenced by daily or monthly variations of RH. Approximately 75% of haze occurs when the RH is exceeds 60% or the PM_2.5 mass concentration is below 105 μg/m³ over China, no matter which haze identification method is used. Haze has obvious regional distribution characteristics and is relatively higher in Beijing-Tianjin-Hebei and its surrounding areas, and the middle and lower reaches of the Yangtze River. The 6-year mean annual total haze identified by the MGB method is 1167 h for mainland China. Compared with MGB, M80 underestimates the haze hour by -34%, and M90 produces a smaller positive overestimation by 18%. The annual total haze hour of China and its three major economic regions shows significant decreasing trends regardless of the identification method used. Daily variation of haze is obtained in this research via automatic visibility measurement. The daily cycles of haze hour identified by MGB and M90 are similar, whereas that identified by M80 behaves differently affected by daily variation of RH. Haze hour is high in winter and low in summer.