• 关键词：PM_2.5浓度  大气污染  全域空间自相关  自然正交函数  时空变化特征

• 基金项目：国家自然科学基金项目（No.41761021）；教育部人文社会科学研究青年基金项目（No.17YJC790061）；中国博士后学基金面上资助项目（No.2017M621740）

• 姜磊
• 浙江财经大学经济学院, 杭州 310018

• 周海峰
• 浙江财经大学经济学院, 杭州 310018

• 赖志柱
• 华东师范大学地理信息科学教育部重点实验室, 上海 200241

• 柏玲
• 南昌大学经济管理学院, 南昌 330031

• 陈忠升
• 西华师范大学国土资源学院, 南充 637002

• 摘要：近年来我国雾霾事件频发.采用2015-2017年全国329个地级及以上城市PM_2.5浓度每小时监测数据，利用全域空间自相关法、自然正交函数和空间描述统计分析的方法，从时空视角来揭示PM_2.5浓度的时间动态变化规律以及空间分布特征.研究发现：①从全国范围内来看，PM_2.5浓度均值逐年降低，降幅最高为夏季，最低为冬季，PM_2.5浓度位于40~60 μg·m^-3之间的城市降幅较大.PM_2.5浓度年内表现为"冬高夏低，春秋居中"的时间动态变化规律，且各年PM_2.5浓度达优良率不断提高.②细颗粒物污染改善程度最大的为德州，京津冀城市群和长三角城市群改善程度居中.全国PM_2.5污染范围逐年缩小，但新疆西部和冀鲁豫仍为高污染区，西南和东南沿海地区为低污染区.各区域污染的空间集聚逐年缩小.优良达标率在空间分布特征上无显著变化.③"大气十条"部分指标已完成，未来细颗粒物污染治理重点区域仍以京津冀地区为核心.在防治空气污染方面，必须加强区域联防联控机制.

• Abstract：Haze-fog events have frequently occurred in China in recent years. Based on a day-level data-set of PM_2.5 concentrations of 329 cities in China from 2015 to 2017, this paper employs the global spatial autocorrelation analysis, empirical orthogonal function, and spatial statistics methods to investigate the temporal dynamics of PM_2.5 concentrations and their spatial patterns from the spatio-temporal perspective. The findings are the following. ① PM_2.5 concentrations are getting decrease year by year nationwide. Specifically, in terms of PM_2.5 concentrations summer season sees the biggest decrease during the sample period and winter season a slight decrease. Moreover, these cities with PM_2.5 concentrations ranging from 40~60 μg·m^-3 see the largest decrease. In other words, air quality of these cities is getting improve. Besides, it shows a significant temporal pattern of high concentrations in winter, low concentrations in summer and intermediate concentrations in spring and autumn. ② Regarding PM_2.5 concentrations of 329 cities, Dezhou city sees the biggest decrease. However, Jing-Jin-Ji urban cluster and Yangtze River Delta urban cluster show slight improvements of air quality during the sample period. On the other hand, the area of air pollution displays an ever-narrowing trend. However, the western Xinjiang and Hebei-Shandong-Henan provinces are still highly-polluted areas while the cities of the southwest and southeast China usually have low PM_2.5 concentrations, indicating better air quality. Besides, spatial clustering of PM_2.5 concentrations is narrowing year by year. Spatial distribution of the excellence rate of air quality of cities shows no remarkable variation. ③ Parts of targets that are set in the air pollution prevention and control action plans have already accomplished. The focus area of air pollution prevention and control should be still on the Jing-Jin-Ji urban cluster. Moreover, the joint efforts should be urgently needed to prevent and control air pollution.

• Key words：PM_2.5 concentration  air pollution  global spatial autocorrelation  empirical orthogonal function  spatio-temporal characteristics

• 
  摘要点击次数： 1454 全文下载次数： 1386