麦盖提县大气颗粒物污染特征及潜在源区分析

Analysis of Atmospheric Particulate Matter Pollution Characteristics and Potential Source Regions in Magaiti County

摘要: 基于全球同化系统(GDAS)数据及2018~2024年麦盖提县环境监测站逐小时颗粒物浓度数据等多源资料，运用大气团后轨迹聚类分析、潜在源贡献因子法(PSFC)和浓度权重轨迹法(CWT)等方法，分析麦盖提县颗粒物浓度的时空变化特征及不同季节外源污染物传输的路径及潜在源区。结果表明：(1) 麦盖提县PM_2.5和PM₁₀污染天数分别为132 d和925 d，年均浓度依次为81.2 μg·m⁻³和271.1 μg·m⁻³，按季节变化来看，不同季节、不同输送路径对麦盖提县PM_2.5和PM₁₀污染物影响的差异显著，PM_2.5浓度表现为春季 > 冬季 > 秋季 > 夏季，PM₁₀浓度表现为春季 > 秋季 > 冬季 > 夏季。(2) 麦盖提县全年，东北方向输送的气流对应的PM_2.5和PM₁₀浓度最大，轨迹占比为25.7%，轨迹途径区域分别为巴楚县和墨玉县等；其次是麦盖提本区域内气流对应的PM_2.5和PM₁₀浓度，轨迹占比为59.4%，移动速度较慢；西北方向输送气流对应的PM_2.5和PM₁₀浓度相对较低，轨迹占比为14.9%，且其输送距离最远，轨迹途径区域分别为乌恰县–疏附县–岳普湖县–麦盖提县。(3) 麦盖提县颗粒物春季PSFC值和CWT值最大，夏季最小。PSFC和CWT值高值区分布在阿克苏–巴楚县–麦盖提县–墨玉县呈带状区域。(4) 由随机森林模型结果表明，PM_2.5受气温和湿度的影响比较大，PM₁₀受到风速和温度的影响比较大。

Abstract: Based on multi-source data such as Global Data Assimilation System (GDAS) data, ERA5 reanalysis data, and hourly particulate matter concentration data from the environmental monitoring station in Maigaiti County from 2018 to 2024, methods such as air mass back-trajectory cluster analysis, Potential Source Contribution Function (PSFC), and Concentration Weighted Trajectory (CWT) were used to analyze the temporal and spatial variation characteristics of particulate matter concentrations in Maigaiti County, as well as the transport paths and potential source areas of exogenous pollutants in different seasons. The results show that: (1) In Maigaiti County, the number of days with PM_2.5 and PM₁₀ pollution was 132 and 925, respectively, with corresponding average annual concentrations of 81.2 μg·m⁻³ and 271.1 μg·m⁻³. Seasonally, the influence of different seasons and transport pathways on these pollutants varied significantly. PM_2.5 concentrations followed the order of spring > winter > autumn > summer, while PM₁₀ concentrations followed spring > autumn > winter > summer. (2) Annually, the highest PM_2.5 and PM₁₀ concentrations in Maigaiti County were associated with airflows from the northeast, which accounted for 25.7% of the trajectories and passed through areas such as Bachu and Moyu counties. Local airflows within Maigaiti contributed the second-highest concentrations, characterized by slower movement and comprising the majority (59.4%) of trajectories. Notably, airflows from the northwest, despite having the longest transport distance and passing through Wuqia, Shufu, and Yuepuhu counties, corresponded to relatively lower PM_2.5 and PM₁₀ levels, representing only 14.9% of the total trajectories. (3) The PSFC and CWT values of particulate matter in Maigaiti County are the largest in winter and the smallest in summer. The high-value areas of PSFC and CWT values are distributed in a strip-shaped area from Aksu to Bachu County, Maigaiti County, and Moyu County. (4) The results of the random forest model show that PM_2.5 is more affected by air temperature and humidity, and PM₁₀ is more affected by wind speed and temperature.