The transition process of aerosol distribution state with spatio-temporal analysis based on MODIS product

Abstract. Atmospheric aerosol can have an impact on the climate through direct and indirect radiative forcing and cause air quality problems to harm human health [Charlson, R.J., et al. 1992]. Many scholars have made a lot of contributions to aerosol concentration analysis, but the analysis of chaotic aerosol concentration time series is always unable to perfectly solve the inherent randomness brought by complicated dynamic behaviors [Abarbanel, H.D.I, 1996]. This research has established the aggregation and dispersion states of aerosol optical thickness in fixed region and fixed time period by calculating spatial statistics of value-containing pixels in remote sensing images. And it has also analyzed the process of this state which is influenced by several meteorological factors. This kind of aggregation and dispersion is a conceptual method of spatial statistics in geography. The research compares the aerosol AOD value and its distribution area in the atmosphere from the signal frequency and utilizes the methods of nonlinear analysis and wavelet analysis to solve for its state transition characteristics at a regular point in time. Then the aerosol state transfer process is unfolded in space to analyze the interaction between the sequential meteorological factors and the aerosol state transfer process, as well as the comparison of the differences and similarities between the aerosol state transfer process in urban areas and non-urban areas. The research can not only provide theoretical foundation for pollution warning on the basis of long time series but also provide a new idea for analyzing the nonlinear dynamic process of the atmospheric system, which is the analysis of the change process of the chaotic system by spatial deploy at a specific time point.