Abstract:

[Objective] In recent years, China has achieved marked and sustained reductions in fine particulate matter (PM_2.5) pollution. In contrast, tropospheric ozone (O₃) has emerged as a growing environmental concern, with ambient concentrations showing a fluctuating upward trend followed by a persistent plateau at elevated levels. As a secondary pollutant formed through photochemical reactions between nitrogen oxides (NO_x) and volatile organic compounds (VOCs)—a process that is highly nonlinear and strongly dependent on meteorological conditions—O₃ poses serious risks to human health, agricultural yields, and terrestrial ecosystems. The Chinese government has recognized this challenge and implemented active measures. The Action Plan for Continuous Improvement of Air Quality, promulgated in 2023, aims to maintain low PM_2.5 concentrations while further reducing NO_x and VOC emissions. The overarching goal is to improve the atmospheric environmental management system and strengthen overall pollution-control capacity, thereby achieving coordinated and sustained improvements in both PM_2.5 and O₃ air quality. Hangzhou, located in northern Zhejiang Province along the lower reaches of the Qiantang River, is a key city in the Yangtze River Delta urban agglomeration. It is characterized by rapid economic growth and sustained population expansion, making it a representative eastern Chinese urban center for investigating the characteristics and photochemical reactivity of ambient VOCs and their contributions to O₃ formation under complex anthropogenic emission conditions. Previous studies of ambient VOCs in Hangzhou have relied largely on long-term observational datasets and have focused on concentration levels, major emission sources, and species dominant in photochemical reactions. However, they have generally overlooked the influence of photochemical loss on measured VOC concentrations, potentially introducing systematic bias into both concentration assessments and source apportionment. To address this gap and better understand the role of VOCs role in summertime O₃ pollution, this study corrected the observed concentrations to provide a more representative characterization of VOC composition and source contributions. [Methods] Hourly data were collected at the Jingfang Middle School monitoring station in Shangcheng District, Hangzhou, from August 6 to 31, 2023. A photochemical age-based parameterization method was used to correct the observed concentrations of 105 VOC species and estimate their initial concentrations prior to atmospheric processing. Positive matrix factorization (PMF) was subsequently applied for source apportionment. In addition, the observation period was divided into O₃ pollution and non-pollution episodes, and the relationships between ambient VOC characteristics and the O₃ pollution were investigated by comparing the observed concentrations with the corresponding estimated initial concentrations. [Results] During the observation period, the hourly mass concentration of O₃ ranged from 6 to 251 μg·m⁻³, with a mean of 97.37 μg·m⁻³, and exhibited a characteristic single-peak diurnal pattern. Notably, from August 11 to 14, the maximum daily 8-hour average O₃ concentration (MDA8 O₃) exceeded 160 μg·m⁻³, surpassing the Grade II limit specified in the Ambient Air Quality Standard. This period was therefore defined as the pollution episode, whereas the remainder of the observation period was classified as the non-pollution episode. Statistical analysis showed that the mean hourly O₃ concentration during the pollution episode was 139.28 μg·m⁻³, approximately 1.55 times that during the non-pollution episode, indicating a significant difference between the two periods. The hourly mean volume fraction of ambient VOCs ranged from 12.01×10⁻⁹ to 67.36×10⁻⁹, with an overall mean of 26.35×10⁻⁹. After photochemical correction, the hourly mean volume fraction increased to 44.11×10⁻⁹, corresponding to a photochemical loss rate of 40.26%. In terms of composition, oxygenated VOCs (OVOCs) and alkanes together accounted for the largest proportions, representing 62.62% of the observed concentrations and 52.70% of the estimated initial concentrations. Alkenes exhibited a concentration loss rate exceeding 80%, consistent with their inherently high photochemical reactivity and underscoring the importance of photochemical correction when evaluating the actual composition and chemical reactivity of ambient VOC mixtures in urban environments. The maximum incremental reactivity (MIR) method was used to quantify the ozone formation potential (OFP) of ambient VOCs. The total initial OFP was 220.71×10⁻⁹, with a loss rate of 78.54%, indicating that a substantial fraction of the most reactive VOC species had already undergone photochemical transformation before reaching the monitoring site. This finding further suggests that OFP estimates based solely on observed concentrations may substantially underestimate the initial photochemical reactivity of VOC emissions. Alkenes were identified as the dominant contributors to OFP because of their higher photochemical reactivity than other functional groups. Although the mean VOC concentration during the O₃ pollution episode was higher than that during the non-pollution episode, the OFP was slightly lower during the pollution episode. This difference was partly attributable to relative humidity. The mean relative humidity during the non-pollution episode was 12.76% higher than that during the pollution episode, and higher humidity may favor the formation of hydroxyl radicals (·OH), thereby accelerating the photochemical loss of VOCs. Source apportionment of the observed VOC concentrations using PMF resolved six source categories: secondary sources, industrial emissions, oil and gas volatilization, solvent coatings, industrial cleaning, and biogenic emissions. After photochemical age correction, five source factors were resolved: combustion sources, industrial emissions, oil and gas volatilization, solvent coatings, and biogenic emissions. In the source apportionment based on observed concentrations, secondary sources were the dominant contributor, accounting for 27.86%; in contrast, combustion sources emerged as the dominant contributor in the source apportionment based on corrected initial concentrations, accounting for 26.52%. [Conclusion] Photochemical correction can effectively reduce bias in PMF source apportionment by estimating initial VOC concentrations, thereby producing more representative factor profiles and more reliable source-contribution estimates by reducing distortions associated with differential photochemical losses among VOC species. These findings demonstrate the importance of incorporating photochemical correction into both scientific research and air-quality management. In the source apportionment results based on both observed and corrected concentrations, industrial emissions, oil and gas volatilization and solvent coatings consistently made substantial contributions and collectively represented the major local anthropogenic emission sources. These sources should therefore by prioritized when developing targeted emission-reduction strategies to improve O₃ air quality in Hangzhou and surrounding areas.

Key words: volatile organic compounds, photochemical loss, ozone pollution, source apportionment, Hangzhou

摘要：

【目的】明确杭州市夏季大气挥发性有机物（VOCs）对臭氧（O₃）污染的影响，为精准防控提供依据。【方法】基于当地夏季观测数据，采用光化学年龄参数法校正105种VOCs的观测体积分数，利用正交矩阵因子分解（PMF）模型进行来源解析。【结果】观测期间O₃小时平均质量浓度为97.37 μg∙m⁻³，污染期均值为139.28 μg∙m⁻³，是非污染期的1.55倍。VOCs平均观测体积分数为26.35×10⁻⁹，初始体积分数为44.11×10⁻⁹；初始臭氧生成潜势（OFP）总量为220.71×10⁻⁹，损耗率为78.54%。含氧VOCs（OVOCs）和烷烃是体积分数优势组分，占观测VOCs的62.62%、初始VOCs的52.70%；烯烃的OFP值最高且体积分数损耗率超过80%。O₃污染期VOCs平均体积分数高于非污染期，但OFP值略低；非污染期烯烃与OVOCs的光化学损耗更大，且平均湿度更高。高湿条件有利于·OH生成，进而加速损耗。PMF模型基于观测数据解析出二次源、工业排放、油气挥发、溶剂涂料、工业清洗和天然源6个污染源；经光化学校正，污染源减少至5个，分别为燃烧源、工业排放、油气挥发、溶剂涂料和天然源，且燃烧源以26.52%的贡献率居于首位。【结论】光化学损耗会掩盖环境VOCs的真实排放特征，应在科学研究与政策制定中引入光化学校正，以准确识别真实排放来源。

关键词: 挥发性有机物, 光化学损耗, 臭氧污染, 来源解析, 杭州