The annual - inter - annual, seasonal and monthly concentration values of nitrogen dioxide in 346 major cities in China, the values of 15 influencing factors and SHAP values.

This study obtained tropospheric nitrogen dioxide (NO₂) concentration data based on Sentinel-5P and employed statistical analysis and local spatial autocorrelation analysis methods to explore the spatiotemporal distribution and variation characteristics of NO₂ concentrations in 346 major cities in China from 2019 to 2023. Using the Random Forest (RF) model and Shapley Additive Explanations (SHAP) values, the influence of 15 factors on ambient NO₂ concentrations was analyzed.The results showed that the NO₂ concentrations in the 346 cities exhibited seasonal variations, with the highest concentrations in winter and the lowest in summer. From 2019 to 2023, NO₂ concentrations showed a fluctuating trend with significant regional differences: the concentrations were relatively higher in provincial capital cities as well as in northern and northeastern China. Temperature-related variables, aerosol optical depth, and leaf area index had significant impacts on NO₂ concentrations. The findings of this study identify key influencing factors and provide a scientific basis for revealing the causes of urban air pollution in China, formulating pollution control strategies, identifying priority areas for remediation, and supporting the development of protection policies.

本研究基于哨兵5P（Sentinel-5P）卫星获取对流层二氧化氮（NO₂）浓度数据，采用统计分析与局域空间自相关分析方法，探究2019-2023年中国346座主要城市的二氧化氮浓度时空分布与变化特征。本研究借助随机森林（Random Forest, RF）模型与夏普利可加解释（Shapley Additive Explanations, SHAP）值方法，分析了15项影响因子对环境二氧化氮浓度的作用效应。研究结果表明，346座城市的二氧化氮浓度呈现显著季节变化特征，冬季浓度最高，夏季浓度最低。2019至2023年间，二氧化氮浓度呈波动变化趋势且区域差异显著：省会城市及中国北方、东北地区浓度相对偏高。与温度相关的变量、气溶胶光学厚度以及叶面积指数对二氧化氮浓度具有显著影响。本研究明确了关键影响因子，可为揭示中国城市大气污染成因、制定污染防控策略、划定污染治理优先区域以及支撑环境保护政策制定提供科学依据。