【摘要】 根据中国环境监测总站发布的2014年5月—2015年4月的中国城市各监测站点细颗粒物(Particulate Matter 2.5,PM2.5)质量浓度小时数据,将161座城市作为节点,以城市间PM2.5质量浓度的相关性与距离的比值作为边的权重,构建了中国城市PM2.5加权网络,并采用Girvan Newman算法(GN算法)对网络进行划分,得到了不同季节中国PM2.5污染的区域分布情况.结果表明,不同季节划分结果的模块性Q函数均在0.7左右,可采用复杂网络对中国城市PM2.5区域进行划分;全年和四季分别划分出7、9、13、6、8个区域,以具体划分结果和连片度衡量中国不同季节的PM2.5污染的区域性程度顺序如下:冬季>秋季>春季>夏季;划分结果与大气污染防治规划提出的"三区六群"范围大致相同,但区域范围在不同季节均存在不同程度的差异.更多还原

【Abstract】 With the hourly Particulate Matter 2. 5( PM2. 5) concentration data from May 2014 to April2015 of 161 cities’ monitoring stations of China from China National Environmental Monitoring Center. Taking the cities as network nodes and the correlation of cities’ PM2. 5concentration divided by its distance as edge weight,weight networks of PM2. 5in cities of China was created,then networks with Girvan Newman algorithm( GN algorithm) was divided and region division result of PM2. 5pollution in different period was got. Results show that the modularity Q of different periods are all around 0. 7 and indicate that a satisfactory region division result of PM2. 5can be got with the complex networks method; 7,9,13,6,8 regions are divided in full year and four seasons,the regional level order in different seasons measured by division results and contiguous degree is winter > autumn > spring > summer; The region division is similar with the range of"Three Region Six Group"proposed from air pollution control plan,but there are different degrees of difference in different seasons of region range.更多还原