Multifractal nature of particulate matters (PMs) in Hong Kong urban air.

In this study, we investigate the persistent variation and the multifractal nature of particulate matter (PM) concentrations from vehicle emissions at a typical traffic intersection of street canyon in Hong Kong. Six size groups of PMs are measured and collected during rush hour sessions on different dates respectively. A recently developed model, namely multifractal detrended fluctuation analysis (MF-DFA), is employed to decompose and analyze the collected database. Through estimating the scaling exponent, it is found that the PM levels from vehicular emissions display long-term correlation characters. By employing MF-DFA method to calculate the generalized Hurst exponent and discuss the multifractal spectrums of all size groups, it is noticed that the fine particulate matters in grain diameter of 0.3-0.499 μm present strong multifractal nature, intensive oscillations of concentration variations, and long-term persistence. For fine particulate matters in the grain diameter ranges from 0.5 μm to 4.99 μm, their similar and weak multifractal natures reflect the self-similarity behaviors among these groups and the gradual decreases of the lasting effects. For large size particulate matters, i.e., grain diameter above 5 μm, certain mono-fractal nature and sharp decay of long-term persistence are obtained, even for intermittent effects. It can therefore be concluded that the fine particulate matter diffuses anomaly and persists for a long time.