京津冀城市群大气污染的时空特征与影响因素解析

京津冀城市群是中国雾霾最严重的区域,在京津冀协同发展背景下,探究该地区大气污染的时空分布和影响因素具有重要意义。运用空间自相关分析和三种空间计量模型,分析了京津冀202个区县PM_2.5的时空分异特征,创新性地对自然与人文影响因素贡献及其空间溢出效应进行系统地甄别和量化。结果表明：2000-2014年来京津冀城市群PM_2.5浓度整体呈上升趋势,季节上呈秋冬高、春夏低,空间上呈东南高、西北低的特点,且城市建成区PM_2.5浓度比周围郊区和农村平均高10~20 μg/m³;2014年仅有13.9%的区县空气质量达标,PM_2.5浓度存在显著的空间集聚性与扩散性,城市间交互影响距离平均为200 km,邻近地区的PM_2.5每升高1%,将导致本地PM_2.5至少升高0.5%;社会经济内因对PM_2.5主要是正向影响,自然外因主要是负向影响;影响因素中对本地大气污染的直接效应贡献强度依次是：年均风速>年均气温>人口密度>地形起伏度>第二产业占比>能源消费>植被覆盖度,人均GDP、年降水量和相对湿度对本地PM_2.5没有显著影响;对邻近地区大气污染具有显著空间溢出效应的因素排序是：植被覆盖度>地形起伏度>能源消费>人口密度;对于自然和人文影响因素应分别采取针对性的适应策略和调控策略,加强区域间联防联控与合作治理,在城市群规划中注重环保规划与立法。

The spatial-temporal characteristics and influencing factors of air pollution in Beijing-Tianjin-Hebei urban agglomeration

Air pollution is now the most serious environmental problem in China, especially for Beijing-Tianjin-Hebei (Jing-Jin-Ji) urban agglomeration. Under the background of Jing-Jin-Ji integrated development, it is of great importance to study the spatio-temporal distribution and its influencing factors. This paper firstly analyzed the spatio-temporal characteristics of PM_2.5 concentration in 202 counties of Jing-Jin-Ji urban agglomeration using spatial analysis and visualization. Then we quantified the effect degree of different natural and human factors and the spatial spillover effect of these factors using Spatial Durbin Model (SDM). The results are as follows: (1) The PM_2.5 concentration was on the rise as a whole from 2000-2014, and it was high in autumn and winter, while low in spring and summer. Seen from spatial distribution, the PM_2.5 concentration was high in the southeast, while low in the northwest of the study area, and the concentration in urban built-up area was 10-20 μg/m³ higher than that of its surrounding suburban and rural areas. (2) The counties, whose PM_2.5 concentration was below 35 μg/m³, only accounted for 13.9% in 2014. There were significant spatial agglomeration and diffusivity of PM_2.5, and the mean distance interacting among the cities was 200 km. The PM_2.5 of a city increased by more than 0.5% for every 1% increase in the average PM_2.5 of neighboring cities. (3) Socio-economic factors have positive impact on PM_2.5, and most of natural factors have negative impact. (4) Among the influencing factors, contributions to local atmospheric pollution with direct effect are: mean wind speed > annual mean temperature > population density > relief amplitude > the secondary industry share of GDP > energy consumption > vegetation coverage; however, per capita GDP, annual precipitation, and relative humidity have no significant effect on local pollution. (5) Contributions to neighborhood atmospheric pollution with spatial spillover effect are: vegetation coverage > relief amplitude > population density. The findings suggest that we should adopt the adaptation strategies for natural factors and control strategies for human factors. A coordinated and strengthened cooperation between local governments should be established for air pollution control, and environmental planning and legislation should be strengthened in a new round of urban agglomeration planning in China.