Asymmetric multiscale behavior in PM2.5 time series: Based on asymmetric MS-DFA

摘要: Abstract Particulate matter with an aerodynamic diameter of 2.5 mm or less (PM2.5) is one the most serious air pollution, considered harmful for people by World Health Organisation. In this paper, we utilized asymmetric multiscale detrended fluctuation analysis (A-MSDFA) method to explore existence correlation properties PM2.5 daily average concentration in two USA cities (Fresno and Los Angeles) Chinese (Hong Kong Shanghai), assess these correlations. The results show existences correlations, degree stronger than that cities. Further, local exponent β ( n ) are smaller 0.5, which indicates anti-persistent long-range time series four addition, reanalyze A-MSDFA secant rolling windows different sizes, can investigate dynamic changes changing window size. Whatever correlations display asymmetries at small scales larger large scales. Moreover, become increasingly weaker increase sizes.