中文关键词PM_2.5时空分布空间聚类分析模糊C均值聚类(FCM)地理时空加权回归 英文关键词PM_2.5spatio-temporal distributionspatial clustering analysisfuzzy-c-means(FCM) clusteringgeographically temporally weighted regression

作者	单位	E-mail
杨文涛	湖南科技大学资源环境与安全工程学院, 湘潭 411201 湖南科技大学地理空间信息技术国家地方联合工程实验室, 湘潭 411201	yangwentao8868@126.com
姚诗琪	香港中文大学地理资源管理学系, 香港 999077	shiqi_yao@outlook.com
邓敏	中南大学地理科学与信息物理学院地理信息系, 长沙 410083
王艳军	湖南科技大学资源环境与安全工程学院, 湘潭 411201 湖南科技大学地理空间信息技术国家地方联合工程实验室, 湘潭 411201

中文摘要 PM_2.5时空分布特征及其与其它污染物的相关关系是PM_2.5时空统计分析的主要研究内容.然而，现有的方法直接从监测站点的角度对时空分布特征进行分析，难以有效地揭示PM_2.5浓度的聚集分布特征；同时，常用的地理加权回归在对PM_2.5与其它污染物间关系进行建模的过程中，缺乏同时考虑时间异质性与空间异质性，从而不能准确地描述依赖关系的时空变异特征.为此，首先借助于空间聚类分析技术，对北京市2014年PM_2.5浓度的聚集结构进行探测，在此基础上，通过聚集结构来分析PM_2.5季节性时空分布特征.然后，利用地理时空加权回归对北京市PM_2.5与PM₁₀季节平均浓度间关系进行建模，依据回归结果分析PM_2.5-PM₁₀间关系的时空变异特征.实验结果表明，春夏季节PM_2.5污染程度及空间变异程度均低于秋冬季节，各季节PM_2.5浓度均表现为北部浓度低、南部浓度高的空间分布特征；地理时空加权回归具有更好的拟合效果，由回归系数进一步可发现，春夏季PM_2.5-PM₁₀相关性低于秋冬季PM_2.5-PM₁₀相关性；各季节均表现为西北部PM_2.5-PM₁₀的相关性高于东南部PM_2.5-PM₁₀的相关性. 英文摘要 Spatio-temporal distribution of PM_2.5 and variations in the relationship between PM_2.5 and other pollutants are the main components of PM_2.5spatio-temporal statistical analysis. Existing methods directly analyze spatio-temporal distribution based on monitoring data; thus, it is difficult to effectively reveal the aggregation structure of PM_2.5 concentrations. Geographically weighted regression, commonly used to model the relationships between PM_2.5 and other pollutants, cannot accurately describe the spatio-temporal variability of dependency. In this study, the clustering structure of PM_2.5 concentrations in Beijing was identified using the spatial clustering algorithm and the seasonal distribution characteristics of PM_2.5 were analyzed based on the clustering results. The relationship between PM_2.5 and PM₁₀ was modeled by geographically and temporally weighted regression and the spatio-temporal variability of dependency was analyzed according to the regression results. The results showed that PM_2.5 pollution levels and spatial variability were lower in spring and summer than those in autumn and winter and the concentration of PM_2.5 in each season was characterized by low spatial distribution in the north and high spatial distribution in the south. Geographically and temporally weighted regression showed better performance; the correlations between PM_2.5 and PM₁₀ in spring and summer are weaker than those in autumn and winter and the correlation between PM_2.5 and PM₁₀ in the northwest is stronger than that in the southeast in each season.

PDF全文下载地址:

https://www.hjkx.ac.cn/hjkx/ch/reader/create_pdf.aspx?file_no=20180225&flag=1&journal_id=hjkx&year_id=2018