基于聚类分析的颗粒物监测网络优化研究

摘要

This study aims to evaluate the performance of cluster analysis to optimize the monitoring network of PM2.5and PM10in Hong Kong.The statistical method was applied to the mass concentration of PM2.5,PM10and the value of PM2.5/PM10,collected in 14 monitoring stationsof Hong Kong from January 2011 to November 2015.The results demonstrate that, for each pollutant, the monitoring stations located in nearby area are characterized by the same specific air pollution characteristics, which would cause a waste of resources.To buildan effective management of air quality monitoring system,the redundant equipment should be transferred to other monitoring stations for allowing further enlargement of the monitored area.Additionally, according to the similar changes of the value of PM2.5/PM10,the concentration of one pollutant can be predicted based on the other one data,this might provid a new mentality for optimizing the allocation of monitoring resource.%为了优化香港环境监测网络,收集香港14个监测站2011年1月1日至2015年11月30日的颗粒物PM2.5、PM10的小时数据进行统计分析.对PM2.5进行聚类,并利用日均浓度变化图进行验证,结果表明,可将监测站分为4类(A、B、C、D类),A类位于城市郊区,B类则位于港口附近,且A、B类的PM2.5日变化特征均呈现双峰型分布,峰值分别出现在09:00和21:00.对PM10进行类似分析结果表明,监测站同样可以分为4类,A类位于九龙区,B类则位于港口附近,而且A、B类的PM10日变化双峰分别出现在11:00和20:00左右.说明污染源头及地形的相似致使某些监测站颗粒物浓度的变化出现相同的趋势,导致监测设备的浪费和管理的冗余.建议建立更高效的空气管理系统,将冗余设备转移到其他地区,扩大空气监控区域.对PM2.5/PM10聚类结果表明,将监测站分为4类,B类均属于路边站,C类则位于居民区.同时还发现同类监测站PM2.5/PM10数值变化相同,并且可以用其中一个站的PM2.5和PM10浓度及另一个站的PM2.5或PM10浓度预测PM2.5或PM10浓度,为优化监测资源提供了一种新的思路.