Spatio-temporal variation and daily prediction of PM2.5 concentration in world-class urban agglomerations of China

PM2.5 exceedances and source appointment as inputs for an early warning system

Between June 2018 and April 2019, a sampling campaign was carried out to collect PM2.5, monitoring meteorological parameters and anthropogenic events in the Sartenejas Valley, Venezuela. We develop a logistic model for PM2.5 exceedances (≥ 12.5 µg m-3). Source appointment was done using elemental composition and morphology of PM by scanning electron microscopy coupled with energy dispersive spectroscopy (SEM-EDS). A proposal of an early warning system (EWS) for PM pollution episodes is presented. The logistic model has a holistic success rate of 94%, with forest fires and motor vehicle flows as significant variables. Source appointment analysis by occurrence of events showed that samples with higher concentrations of PM had carbon-rich particles and traces of K associated with biomass burning, as well as aluminosilicates and metallic elements associated with resuspension of soil dust by motor-vehicles. Quantitative source appointment analysis showed that soil dust, garbage burning/marine aerosols and wildfires are three majority sources of PM. An EWS for PM pollution episodes around the Sartenejas Valley is proposed considering the variables and elements mentioned.

Associations of Fine Particulate Matter Constituents with Metabolic Syndrome and the Mediating Role of Apolipoprotein B: A Multicenter Study in Middle-Aged and Elderly Chinese Adults

Fine particulate matter (PM_2.5) was reported to be associated with metabolic syndrome (MetS), but how PM_2.5 constituents affect MetS and the underlying mediators remains unclear. We aimed to investigate the associations of long-term exposure to 24 kinds of PM_2.5 constituents with MetS (defined by five indicators) in middle-aged and elderly adults and to further explore the potential mediating role of apolipoprotein B (ApoB). A multicenter study was conducted by recruiting subjects (n = 2045) in the Beijing-Tianjin-Hebei region from the cohort of Sub-Clinical Outcomes of Polluted Air in China (SCOPA-China Cohort). Relationships among PM_2.5 constituents, serum ApoB levels, and MetS were estimated by multiple logistic/linear regression models. Mediation analysis quantified the role of ApoB in "PM_2.5 constituents-MetS" associations. Results indicated PM_2.5 was significantly related to elevated MetS prevalence. The MetS odds increased after exposure to sulfate (SO₄^2-), calcium ion (Ca²⁺), magnesium ion (Mg²⁺), Si, Zn, Ca, Mn, Ba, Cu, As, Cr, Ni, or Se (odds ratios ranged from 1.103 to 3.025 per interquartile range increase in each constituent). PM_2.5 and some constituents (SO₄^2-, Ca²⁺, Mg²⁺, Ca, and As) were positively related to serum ApoB levels. ApoB mediated 22.10% of the association between PM_2.5 and MetS. Besides, ApoB mediated 24.59%, 50.17%, 12.70%, and 9.63% of the associations of SO₄^2-, Ca²⁺, Ca, and As with MetS, respectively. Our findings suggest that ApoB partially mediates relationships between PM_2.5 constituents and MetS risk in China.

Two-stage deep learning hybrid framework based on multi-factor multi-scale and intelligent optimization for air pollutant prediction and early warning

Effective prediction of air pollution concentrations is of great importance to both the physical and mental health of citizens and urban pollution control. As one of the main components of air pollutants, accurate prediction of PM_2.5 can provide a reference for air pollution control and pollution warning. This study proposes an air pollutant prediction and early warning framework, which innovatively combines feature extraction techniques, feature selection methods and intelligent optimization algorithms. First, the PM_2.5 sequence is decomposed into several subsequences using the complete ensemble empirical mode decomposition with adaptive noise, and then the new components of the subsequences with different complexity are reconstructed using fuzzy entropy. Then, the Max-Relevance and Min-Redundancy method is used to select the influencing factors of the different reconstructed components. Then, a two-stage deep learning hybrid framework is constructed to model the prediction and nonlinear integration of the reconstructed components using a long short-term memory artificial neural network optimized by the gray wolf optimization algorithm. Finally, based on the proposed hybrid prediction framework, effective prediction and early warning of air pollutants are achieved. In an empirical study in three cities in China, the prediction accuracy, warning accuracy and prediction stability of the proposed hybrid framework outperformed the other comparative models. The analysis results indicate that the developed hybrid framework can be used as an effective tool for air pollutant prediction and early warning.