• 关键词：PM_2.5  广州  时空分布  系统聚类法  经验正交函数（EOF）

• 基金项目：广东省科技计划项目（科技创新平台类）（No.2019B121201002）；广州市科技计划项目（No.202002020065，202102080679）

• 陈彦宁
• 广东省广州生态环境监测中心站，广州 510006

• 张金谱
• 广东省广州生态环境监测中心站，广州 510006;中国科学院广州地球化学研究所，有机地球化学国家重点实验室，广东省环境资源利用与保护重点实验室，广州 510640;中国科学院大学，北京 100049

• 裴成磊
• 广东省广州生态环境监测中心站，广州 510006;中国科学院广州地球化学研究所，有机地球化学国家重点实验室，广东省环境资源利用与保护重点实验室，广州 510640;中国科学院大学，北京 100049

• 邱晓暖
• 广东省广州生态环境监测中心站，广州 510006

• 陈漾
• 广东省广州生态环境监测中心站，广州 510006

• 摘要：对大气污染物进行时空分布特征研究是开展大气污染防治的关键技术支撑.本研究基于广州市52个城市环境空气质量监测站点数据，采用系统聚类法、经验正交函数 （EOF）等方法分析了2016—2020年广州市PM_2.5浓度的时空分布特征.结果表明：①2016—2020年广州市PM_2.5污染改善显著，PM_2.5年均浓度从35.9 μg·m^-3下降至23.0 μg·m^-3，达标比例由96.2%上升至100%；PM_2.5干季平均浓度为湿季的1.54倍， 国控点超标天数为湿季的10.5倍；PM_2.5浓度日变化曲线峰谷值浓度差由7.5 μg·m^-3下降至3.9 μg·m^-3，日变化幅度趋于平缓.②广州市PM_2.5浓度最高值区主要分布在东西两侧，高值区域范围逐年减小，全市PM_2.5浓度分布趋于均匀；采用系统聚类法可将广州市PM_2.5分成北部、中北部、 南部、中南部4个污染区，其中，北部区PM_2.5浓度下降率仅为其他污染区的1/2，推测其PM_2.5浓度下降可能更多地由区域背景浓度的下降贡献；EOF分解前3模态累积方差贡献率达93%，分别可表征PM_2.5总体污染程度、在南北方向上的区域输送特征及由外围区域向中心城区聚集的 污染特征.

• Abstract：Research on the spatial and temporal distribution characteristics of air pollutants is the key technical support for air pollution prevention and control. Based on the data of 52 urban ambient air quality monitoring stations in Guangzhou， this study used systematic clustering and empirical orthogonal function （EOF） methods to analyze the temporal and spatial distribution characteristics of PM_2.5. The results show that： ① the PM_2.5 pollution in Guangzhou was significantly improved from 2016 to 2020， and the annual average concentration of PM_2.5 decreased from 35.9 μg·m^-3 to 23.0 μg·m^-3. The percentage of attaining the national ambient air quality standard has increased from 96.2% to 100%. The average concentration of PM_2.5 in the dry season is 1.54 times higher than that of the wet season， and the number of days exceeding the national ambient air quality standard at the national control sites is 10.5 times higher than that in the wet season. The peak-to-valley concentration difference of the diurnal variation curve of PM_2.5 concentration decreased from 7.5 μg·m^-3 to 3.9 μg·m^-3， and the diurnal variations tended to be gentle. ② The highest PM_2.5 concentration areas in Guangzhou are mainly distributed in the east and west， and the range of high-value areas decreases year by year， and the distribution of PM_2.5 concentration in the city tends to be even. Using the systematic clustering method， PM_2.5 in Guangzhou city can be divided into four pollution zones： northern， north-central， southern and south-central. The decrease rate of PM_2.5 concentration in the northern zone is only half of that in other pollution zones， and it is presumed that the decrease of its PM_2.5 concentration may be more contributed by the decrease of regional background concentration. The cumulative variance contribution of the first three modes of EOF decomposition reaches 93%， which can characterize the overall PM_2.5 pollution level， the regional transport characteristics in the north-south direction and the pollution characteristics of PM_2.5 gathering from the peripheral areas to the central city， respectively.

• Key words：PM_2.5  Guangzhou  spatial and temporal distribution  systematic clustering method  empirical orthogonal function（EOF）

• 
  摘要点击次数： 162 全文下载次数： 246