Huang Yan, Wang Qingqing, Ou Xiaojie, Sheng Dongping, Yao Shengdong, Wu Chengzhi, Wang Qiaoli
Ecological Environmental Monitoring Station of Deqing County, Huzhou, 313200, China.
College of Environment, Zhejiang University of Technology, Hangzhou, 310032, China.
Heliyon. 2024 Aug 14;10(16):e36303. doi: 10.1016/j.heliyon.2024.e36303. eCollection 2024 Aug 30.
The pursuit of enhanced scientific, refined, and precise ozone and air quality control continues to pose significant challenges. Using data visualization techniques and random forest (RF) algorithms, the temporal distribution of atmospheric pollutants and the interrelationship between O concentration and its influential factors were investigated with one-year monitoring data in Deqing county in 2021. The local atmospheric conditions predominantly belonged to NOx-sensitive and transition zone. Extremely high O concentration were primarily observed when temperatures (T) exceeded 30 °C, with relative humidity (RH) ranging between 30 and 60 %. NO, RH and T were identified as the top 3 important factors, and O concentration have stronger linearly relationship to RH and T, while stronger nonlinearly relationship to NO. By employing an optimized RF model, controlling consistent mild and high reaction atmospheric conditions, the O concentration response to the change of individual influencing factors was acquired. The O concentration increased and then decreased in response to the increasing NO concentration, displaying a characteristic inflection point at 10 μg m. More reactive radicals produced at higher VOCs concentration and continuing NO cycle at lower NO concentration, resulting in the acceleration in the direction of producing more O. Therefore, the significant different O response to variation of VOCs and NO concentration between mild and high reaction atmospheric conditions, as well as the existing of oxidant elevation should be considered in local air quality control. This study demonstrates the efficacy of ML methods in simulating nonlinear response of O, supports the understanding of local O formation and quick guidance for precise local O pollution control and the related strategies.
追求更科学、精细和精确的臭氧及空气质量控制仍然面临重大挑战。利用数据可视化技术和随机森林（RF）算法，结合德清县2021年的一年监测数据，研究了大气污染物的时间分布以及臭氧浓度与其影响因素之间的相互关系。当地大气条件主要属于对氮氧化物敏感和过渡区域。臭氧浓度极高主要出现在温度超过30°C且相对湿度在30%至60%之间时。一氧化氮、相对湿度和温度被确定为最重要的三个因素，臭氧浓度与相对湿度和温度呈较强的线性关系，而与一氧化氮呈较强的非线性关系。通过采用优化的随机森林模型，控制温和及高反应性大气条件的一致性，获得了臭氧浓度对各个影响因素变化的响应。随着一氧化氮浓度的增加，臭氧浓度先增加后降低，在10μg/m处呈现出特征拐点。在较高的挥发性有机化合物浓度下产生更多活性自由基，在较低的一氧化氮浓度下持续一氧化氮循环，导致向产生更多臭氧的方向加速。因此，在当地空气质量控制中应考虑温和及高反应性大气条件下臭氧对挥发性有机化合物和一氧化氮浓度变化的显著不同响应，以及氧化剂升高的情况。本研究证明了机器学习方法在模拟臭氧非线性响应方面的有效性，有助于理解当地臭氧的形成，并为精确的当地臭氧污染控制及相关策略提供快速指导。
