基于空间插值的西安市重空气污染期间主要污染物时空变化特征及相关性分析

西安市是我国承东启西、连接南北的战略性枢纽城市，但其长期受到重空气污染的影响.基于2018年11月24日-12月3日西安市及其周边7个地级市共38个环境质量监测站点的逐时数据，利用空间插值、趋势分析和相关性分析方法，研究了西安市一次重空气污染期间六大污染物（PM_2.5、PM₁₀、CO、NO₂、SO₂和O₃）的质量浓度时空变化及彼此间的相关关系.结果表明:①IDW（inverse distance weighting，反距加权插值法）和OKri（ordinary Kriging，普通克里格插值法）均能较好地获得西安市空气污染物的时空变化情况，但IDW的插值精度优于OKri，距离指数为7的IDW可以满足西安市空气污染物时空变化模拟的要求.②研究期间，西安市首要污染物为PM_2.5和PM₁₀，二者分别是中度-重度污染及严重-"爆表"污染天气的首要贡献因子.③ρ（PM_2.5）、ρ（PM₁₀）、ρ（CO）、ρ（NO₂）和ρ（SO₂）均呈中部高、两边低，北部高、南部低的空间分布特点，而ρ（O₃）则相反；PM_2.5、PM₁₀、O₃污染程度日趋严重，NO₂污染程度逐渐缓解.④ρ（PM_2.5）、ρ（NO₂）、ρ（CO）之间呈中等正相关，三者在时空变化上具有较高的一致性；ρ（SO₂）与ρ（PM_2.5）、ρ（NO₂）、ρ（CO）均呈弱正相关；ρ（O₃）与ρ（NO₂）、ρ（CO）均呈弱负相关.受扬尘天气和特殊风向及地形共同影响，西安市PM₁₀出现"爆表"现象，导致ρ（PM₁₀）与其他污染物质量浓度之间的相关性不明显.研究显示，距离指数为7的IDW适合西安市空气污染情况时空变化的模拟，重污染天气条件下，西安市ρ（PM_2.5）、ρ（NO₂）、ρ（CO）之间具有较高的同源性，但各污染物间时空变化和相关性关系较复杂. 

Spatiotemporal Characteristics and Relationships of Main Air-Pollutants during a Typical Heavy Air Pollution in Xi'an City Based on a Spatial Interpolation Method

Based on the hourly observations of air quality monitor stations during November 24^th to December 3^rd, 2018 of Xi''an City and 7 cities surrounding it, we studied the spatiotemporal characteristics of the main air pollutant concentrations (PM_2.5, PM₁₀, CO, NO₂, SO₂ and O₃) and their interrelationships during a typical heavy air pollution event in Xi''an City by means of spatial interpolation method, trend and correlation analysis. The results showed:(1) Both Inverse Distance Weight interpolation (IDW) method and the Ordinary Kriging Interpolation method could obtain the spatiotemporal distributions and variations of the air-pollutants'' concentrations, but the former performed better. The IDW with a distance index of 7 was picked as the best interpolation method for Xi''an City in the end. (2) The main air-pollutants of Xi''an City during the ten days were PM_2.5 and PM₁₀, which were the primary contributing factors of moderate to severe air pollution and severe to extreme air pollution, respectively. (3) The concentrations of the first five pollutants were relatively high in the north and/or middle of Xi''an City, but low in the south and/or out sides of Xi''an City, which was just the opposite of O₃ concentration. Besides, the concentrations of PM_2.5, PM₁₀ and O₃ showed increasing trends throughout Xi''an City during the ten days, and the inverse of NO₂ was true. (4) PM_2.5, NO₂ and CO were positively correlated, and the spatial-distribution patterns and temporal trends of these three air-pollutants showed high consistency and homology. SO₂ had weak positive correlations with PM_2.5, NO₂ and CO, while O₃ had weak negative correlations with NO₂ and CO. SO₂ weakly but positively correlated with PM_2.5, NO₂ and CO. PM₁₀ had little and insignificant correlations with all the other air-pollutants because the extremely heavy air-pollution led by PM₁₀ disturbed the normal interactions between PM₁₀ and other air-pollutants. In this heavy air pollution event, high degree of homology exhibited in the variations of PM_2.5, NO₂ and CO, but typical attentions should be paid to distinguish the complicated correlations between PM₁₀ and other air-pollutants.