基于Moran’s I指数的京津冀地区PM2.5空间自相关分析

选取2014~2018年京津冀地区环境监测站点PM_2.5浓度观测数据,基于Moran's I指数,分析了该地区PM_2.5浓度的时空变化特征以及空间相关性。研究表明:2014~2018年京津冀地区PM_2.5浓度空间分布差异显著,2014年良、轻度、中度污染面积约占1/3,降至2018年以“良”为主;PM_2.5浓度空间分布北部、西部优于南部、东部地区,山区优于平原地区;PM_2.5浓度年平均值、IQR、Q1、Q3值及标准差均呈下降趋势,达到且接近于“良”以上程度的站点,从2014年2成增长至2018年8成;通过对历年及各月PM_2.5浓度全局Moran's I指数分析,I值(除7月外)均超过0.5,生成的P值均小于显著水平0.01,Z得分均超过了临界值2.58;年、月平均值均呈显著的集聚效应,年值集聚效应呈减弱趋势,月值集聚效应秋、冬季大于春、夏季,空间集聚效应与污染程度呈显著正相关;局部空间自相关历年、各月H-H集聚主要出现在京津冀南部地区,L-L集聚主要出现在北部地区。京津冀地区PM_2.5浓度值分布存在空间异质性,主要影响因素为地理、气候和经济环境。

Spatial Autocorrelation Analysis of PM2.5in Beijing-Tianjin-Hebei Region Based on Moran's I Index

The observation data of PM_2.5concentration of environmental monitoring stations in the Beijing-Tianjin-Hebei region from 2014 to 2018 were selected and analyzed based on Moran's I index.According to the research,the spatial distribution of PM_2.5concentration in the Beijing-Tianjin-Hebei region from 2014 to 2018 was significantly different.In 2014,the area with good,mild and moderate pollution accounted for about one-third of the total area,which decreased to"good"in 2018.The spatial distribution of PM_2.5concentration in the north and west was better than that in the south and east,and that in the mountainous area was better than that in the plain area.The annual average,IQR,Q1,Q3 and standard deviation of PM_2.5concentration all showed a downward trend,and the number of sites reaching or approaching the degree of"good"or above increased from 20%in 2014 to 80%in 2018.Through the analysis of the global Moran's I index of PM_2.5concentration over the years and in each month,the I value(except July)was over 0.5,the P value generated was less than 0.01,and the Z score was all above the critical value of 2.58.The annual and monthly mean values showed significant clustering effect,and the annual clustering effect showed a weakening trend.The monthly clustering effect was greater in autumn and winter than that in spring and summer,and the spatial clustering effect was significantly positively correlated with the pollution degree.Local spatial autocorrelation H-H cluster mainly appeared in the southern region of Beijing-Tianjin-Hebei in the years and months,and L-L cluster mainly appeared in the northern region.Spatial heterogeneity exists in the distribution of PM_2.5concentration distribution in the Beijing-Tianjin-Hebei region,and the main influencing factors were geography,climate and economic environment.