【摘要】 为实现空气质量的有效监控，从复杂网络的视角研究PM_2.5的相互作用，在传统复杂网络的基础上，采用格兰杰因果检验方法确定京津冀晋鲁豫地区城市间PM_2.5作用关系，引入生物学中的营养一致性理论，探索PM_2.5网络层次特征，从经济社会因素和气象因素两种视角，对京津冀晋鲁豫地区PM_2.5复杂网络特征进行归因分析。结果发现：京津冀晋鲁豫地区PM_2.5具有显著的空间网络特征，区域间的相互作用关系和传输关系复杂，不同季节之间PM_2.5网络的稳定性不同；相比于高中介中心性节点，城市春秋两季网络的效率对度中心性高的节点更敏感，冬季则对二者的敏感程度相近；北京、郑州、济南、太原4个代表城市的PM_2.5浓度与经济社会因素之间有不同程度的相关关系，与气象因素之间则存在不同强度的长程正相关性特征。更多还原

【Abstract】 To realize effective monitoring of air quality, this research studied the interaction of PM_2.5 from the perspective of complex network. Based on the traditional complex network, the research used Granger causality test to determine the influence relationship of PM_2.5 among cities in Beijing-Tianjin-Hebei-Shanxi-Shandong-Henan regions, and introduce the theory of trophic coherence in biology to explore the hierarchical characteristics of PM_2.5 network. From the perspectives of both economic and social factors and meteorological factors, the research conducted the attribution analysis of the characteristics on the complex network of PM_2.5 in Beijing-Tianjin-Hebei-Shanxi-Shandong-Henan regions. The results show that PM_2.5 in Beijing-Tianjin-Hebei-Shanxi-Shandong-Henan regions has significant spatial network characteristics, and the regional interaction and transmission relationship are complex, while the stability of PM_2.5 network varies by season. Compared with nodes with high betweenness centrality, the efficiency of urban network in spring and autumn is more sensitive to nodes with high degree centrality, while in winter, the degree of sensitivity to both is similar. The concentrations of PM_2.5 in Beijing, Zhengzhou, Jinan and Taiyuan are correlated with economic and social factors to varying degrees, and there are long-range positive correlations of varying strengths with meteorological factors.更多还原