Multifractal property and long-range cross-correlation behavior of particulate matters at urban traffic intersection in Shanghai

Abstract

At urban road intersection, the levels of particulate matters within different size groups present multi-variable relationships and have been attracting increasing attentions. In this study, we attend to apply the recently developed multifractal detrended cross-correlation analysis method to investigate the fractal property and the cross-correlation behavior among the non-stationary time series of particulate matters. Six groups of particulate matters with different sizes are measured at a typical traffic intersection of street canyon under different seasons and weather conditions. Based on the collected database, the statistical analyses are carried out and the results indicate close relationships among these groups. Then the multifractal detrended cross-correlation analyses are performed to explore the interaction patterns among all groups, i.e., the fine particulate matters group of 0.3–0.49 cm with other groups, the coarse particulate matters group of 10 μm above with other groups, and the groups of particulate matters between the sizes of 0.5–9.99 μm respectively. In terms of the results, the multifractal property and long-range cross-correlation behavior are observed among all pairs. Comparing to coarse particulate matters, the distinct multifractal spectrum between fine particulate matters with other size groups are observed, which imply that the relevant cross-correlation behaviors are stronger than that in coarse group. It is also found that the cross-correlation behaviors between fine particulate matters with other size groups are highly dependent on the weather conditions while the cross-correlation behaviors between coarse particulate matters with others tend to more depend on the season variations. Finally, the long-range cross-correlation behaviors between them are also confirmed with randomly shuffled series of the observed particulate matters.