Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102728
Yan Jiang , Deyan Wu , Yuqi Guo , Jia Xu , Hongjuan Liu , Aifeng Jia , Chen Li , Duan Ju , Liqiong Guo , Xueli Yang , Qiang Zhang , Bin Han , Zhipeng Bai , Weicheng Chen , Liwen Zhang
Previous studies have linked PM2.5 exposure to adverse pregnancy outcome (APO). However, the specific components responsible remain unidentified, and the potential mediating role of oxidative stress in this relationship is unclear. A cross-sectional study of 431 subjects was conducted in Tianjin, China. Exposure of PM2.5 and its components were obtained through the Tracking Air Pollution in China (TAP) database. Personal blood samples were collected to analyze oxidative stress markers by enzyme-linked immunosorbent assay (ELISA) kits. Logistic regression was used to examine the association between pollutant exposure and APO, while mediation analysis evaluated the role of oxidative stress markers. Each 1 μg/m3 increase in PM2.5 and its components (NH4+, NO3−, SO42−) during the second trimester increased APO risk by 6 % (OR: 1.06, 95 % CI: 1.01, 1.11), 33 % (OR: 1.33, 95 % CI: 1.00, 1.75), 18 % (OR: 1.18, 95 % CI: 1.01, 1.39), and 31 % (OR: 1.31, 95 % CI: 1.02, 1.69), respectively. In the third trimester, these risks increased by 60 % (OR: 1.60, 95 % CI: 1.21, 2.12), 32 % (OR: 1.32, 95 % CI: 1.12, 1.56), and 60 % (OR: 1.60, 95 % CI: 1.22, 2.10), respectively. 3-Nitrotyrosine mediated the effects of BC throughout pregnancy and NH4+, NO3−, and SO42− in the third trimester on LBW, with mediating effects of 48.1 %, 27.3 %, 27.2 %, and 28.9 %, respectively. These findings suggest that protein oxidation mediates the association between NH4+, NO3−, SO42− and APO, with the second and third trimesters identified as critical exposure windows.
{"title":"Exploring the mediating role of oxidative stress in the relationship between PM2.5 components and adverse pregnancy outcomes: Impacts on nucleic acids, proteins, and lipids","authors":"Yan Jiang , Deyan Wu , Yuqi Guo , Jia Xu , Hongjuan Liu , Aifeng Jia , Chen Li , Duan Ju , Liqiong Guo , Xueli Yang , Qiang Zhang , Bin Han , Zhipeng Bai , Weicheng Chen , Liwen Zhang","doi":"10.1016/j.apr.2025.102728","DOIUrl":"10.1016/j.apr.2025.102728","url":null,"abstract":"<div><div>Previous studies have linked PM<sub>2.5</sub> exposure to adverse pregnancy outcome (APO). However, the specific components responsible remain unidentified, and the potential mediating role of oxidative stress in this relationship is unclear. A cross-sectional study of 431 subjects was conducted in Tianjin, China. Exposure of PM<sub>2.5</sub> and its components were obtained through the Tracking Air Pollution in China (TAP) database. Personal blood samples were collected to analyze oxidative stress markers by enzyme-linked immunosorbent assay (ELISA) kits. Logistic regression was used to examine the association between pollutant exposure and APO, while mediation analysis evaluated the role of oxidative stress markers. Each 1 μg/m<sup>3</sup> increase in PM<sub>2.5</sub> and its components (NH<sub>4</sub><sup>+</sup>, NO<sub>3</sub><sup>−</sup>, SO<sub>4</sub><sup>2−</sup>) during the second trimester increased APO risk by 6 % (OR: 1.06, 95 % CI: 1.01, 1.11), 33 % (OR: 1.33, 95 % CI: 1.00, 1.75), 18 % (OR: 1.18, 95 % CI: 1.01, 1.39), and 31 % (OR: 1.31, 95 % CI: 1.02, 1.69), respectively. In the third trimester, these risks increased by 60 % (OR: 1.60, 95 % CI: 1.21, 2.12), 32 % (OR: 1.32, 95 % CI: 1.12, 1.56), and 60 % (OR: 1.60, 95 % CI: 1.22, 2.10), respectively. 3-Nitrotyrosine mediated the effects of BC throughout pregnancy and NH<sub>4</sub><sup>+</sup>, NO<sub>3</sub><sup>−</sup>, and SO<sub>4</sub><sup>2−</sup> in the third trimester on LBW, with mediating effects of 48.1 %, 27.3 %, 27.2 %, and 28.9 %, respectively. These findings suggest that protein oxidation mediates the association between NH<sub>4</sub><sup>+</sup>, NO<sub>3</sub><sup>−</sup>, SO<sub>4</sub><sup>2−</sup> and APO, with the second and third trimesters identified as critical exposure windows.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102728"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102730
Ting Shi, Bo Zhou, Ailin Qi, Kai Wang, Yunpeng Ao, Chenyi Li, Yuling Zhang
Infectious viruses continue to cause severe impacts on the global economy and public health. Although existing studies have demonstrated that meteorological factors and air pollutants are related to virus transmission and mortality, most of them are limited to analyses of specific regions or single factors, often using correlation methods that are not suitable for nonlinear data. Therefore, this study proposes a causal analysis framework based on Convergent Cross-Mapping (CCM) to systematically explore the relationships between environmental factors and the confirmed cases and mortality rates of infectious diseases. First, the simplex projection method was applied to determine the optimal parameters for constructing a causal relationship analysis model. Using COVID-19 as a representative infectious disease, multiple environmental factors and their causal relationships across 39 cities worldwide were investigated. The results show that temperature and humidity consistently ranked first and second among all factors in terms of causal coefficients. Moreover, the causal relationships between humidity, , and and new confirmed cases increased with higher annual averages, while the causal relationships between temperature, , , , and and new deaths became more significant as the annual averages increased. This study validates, on a global scale, the nonlinear causal links between environmental factors and COVID-19, thereby providing new empirical evidence for epidemic prevention, urban environmental governance, and public health policymaking.
{"title":"Causal links between pandemics and weather and air pollution in 39 cities, 2020–2021","authors":"Ting Shi, Bo Zhou, Ailin Qi, Kai Wang, Yunpeng Ao, Chenyi Li, Yuling Zhang","doi":"10.1016/j.apr.2025.102730","DOIUrl":"10.1016/j.apr.2025.102730","url":null,"abstract":"<div><div>Infectious viruses continue to cause severe impacts on the global economy and public health. Although existing studies have demonstrated that meteorological factors and air pollutants are related to virus transmission and mortality, most of them are limited to analyses of specific regions or single factors, often using correlation methods that are not suitable for nonlinear data. Therefore, this study proposes a causal analysis framework based on Convergent Cross-Mapping (CCM) to systematically explore the relationships between environmental factors and the confirmed cases and mortality rates of infectious diseases. First, the simplex projection method was applied to determine the optimal parameters for constructing a causal relationship analysis model. Using COVID-19 as a representative infectious disease, multiple environmental factors and their causal relationships across 39 cities worldwide were investigated. The results show that temperature and humidity consistently ranked first and second among all factors in terms of causal coefficients. Moreover, the causal relationships between humidity, <span><math><mrow><msub><mi>O</mi><mn>3</mn></msub></mrow></math></span>, and <span><math><mrow><msub><mrow><mi>P</mi><mi>M</mi></mrow><mn>2.5</mn></msub></mrow></math></span> and new confirmed cases increased with higher annual averages, while the causal relationships between temperature, <span><math><mrow><msub><mrow><mi>N</mi><mi>O</mi></mrow><mn>2</mn></msub></mrow></math></span>, <span><math><mrow><msub><mrow><mi>P</mi><mi>M</mi></mrow><mn>10</mn></msub></mrow></math></span>, <span><math><mrow><msub><mrow><mi>P</mi><mi>M</mi></mrow><mn>2.5</mn></msub></mrow></math></span>, and <span><math><mrow><mtext>CO</mtext></mrow></math></span> and new deaths became more significant as the annual averages increased. This study validates, on a global scale, the nonlinear causal links between environmental factors and COVID-19, thereby providing new empirical evidence for epidemic prevention, urban environmental governance, and public health policymaking.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102730"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Particulate matter, particularly at sizes less than 2.5 μm (PM2.5), are known to affect respiratory diseases (RSD), but little is explored for their impacts from photochemical alterations. We aimed to examine the association between PM2.5 exposure and RSD to assess correlations in three areas with different exposures scenarios in central Taiwan: P1 (near a coal-powered plant), P2 (downwind at a mountain range base), and P3 (further downwind in a remote mountain valley). Clinical visits (CV) for RSD (ICD9 code 460–496) from 2000 to 2018 were obtained from the National Health Insurance Research Database. PM2.5 concentrations were estimated by utilizing the Hybrid Kriging-Land Use regression model with an XGBoost algorithm. We employed multiple regression analysis to assess the association between PM2.5 concentration and the rate of CV for RSD. We obtained 1,952,413 medical records for analyses. The trends of PM2.5 concentration and the rate of CV for ARI (acute respiratory infections) both decreased during the studied period. More importantly, the rate of CV for ARI was positively associated with the PM2.5 concentration, particularly in the downwind areas (P2 and P3) where photochemical alterations are more significant. In brief, our results showed that PM2.5 was associated with an increase in the rate of CV for RSD, with elevated effects in downwind areas, supporting the potential role of photochemical modifications. Despite a decline in PM2.5 levels, health concerns persist. Future studies considering the overall oxidation capacity of PM to better understand its health impact are recommended.
{"title":"Geographical effects on ambient air pollution and clinical visits for respiratory diseases: a case study with three exposure scenarios","authors":"Hui-Ju Wen , Shu-Li Wang , Chih-Da Wu , Mao-Chang Liang","doi":"10.1016/j.apr.2025.102720","DOIUrl":"10.1016/j.apr.2025.102720","url":null,"abstract":"<div><div>Particulate matter, particularly at sizes less than 2.5 μm (PM<sub>2.5</sub>), are known to affect respiratory diseases (RSD), but little is explored for their impacts from photochemical alterations. We aimed to examine the association between PM<sub>2.5</sub> exposure and RSD to assess correlations in three areas with different exposures scenarios in central Taiwan: P1 (near a coal-powered plant), P2 (downwind at a mountain range base), and P3 (further downwind in a remote mountain valley). Clinical visits (CV) for RSD (ICD9 code 460–496) from 2000 to 2018 were obtained from the National Health Insurance Research Database. PM<sub>2.5</sub> concentrations were estimated by utilizing the Hybrid Kriging-Land Use regression model with an XGBoost algorithm. We employed multiple regression analysis to assess the association between PM<sub>2.5</sub> concentration and the rate of CV for RSD. We obtained 1,952,413 medical records for analyses. The trends of PM<sub>2.5</sub> concentration and the rate of CV for ARI (acute respiratory infections) both decreased during the studied period. More importantly, the rate of CV for ARI was positively associated with the PM<sub>2.5</sub> concentration, particularly in the downwind areas (P2 and P3) where photochemical alterations are more significant. In brief, our results showed that PM<sub>2.5</sub> was associated with an increase in the rate of CV for RSD, with elevated effects in downwind areas, supporting the potential role of photochemical modifications. Despite a decline in PM<sub>2.5</sub> levels, health concerns persist. Future studies considering the overall oxidation capacity of PM to better understand its health impact are recommended.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102720"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102721
Qiaoli Wang , Xiaojie Ou , Chengzhi Wu , Shengdong Yao , Shenlin Huang , Chengcheng Zhu , Ziyi Liao , Kanghui Wang , Shihan Zhang , Jianmeng Chen
Accurate source apportionment of volatile organic compounds (VOCs) serves as a critical foundation for developing targeted control strategies against complex air pollution. While the Positive Matrix Factorization (PMF) model has been extensively employed in pollution source identification, it exhibits inherent limitations when addressing long-term complex pollution scenarios, particularly manifested through slow convergence rates and unavoidable subjective biases during factor resolution. To address these technical challenges, this study innovatively proposes a machine learning-integrated PMF framework (kR-PMF), achieving effective synergy between machine learning algorithms and receptor modeling. The developed kR-PMF model demonstrated superior performance and improved R2 values (0.76–0.95) between simulated and monitored data, indicating enhanced simulation stability. Source apportionment results revealed differential contributions from six major emission categories under PMF and kR-PMF frameworks: biogenic emission sources (8.6 % vs 13.7 %), solvent use (16.5 % vs 14.2 %), industrial emissions (13.6 % vs 10.8 %), gasoline volatilization (14.4 % vs 10.9 %), vehicle emissions (11.8 % vs 7.2 %), and combustion emissions (13.3 % vs 11.1 %). Notably, kR-PMF successfully resolved the secondary source formation process that accounts for 12.5 % of the total volatile organic compounds, a key component that is often obscured in traditional analyses. This methodological breakthrough establishes a robust framework for high-precision VOC source characterization, providing essential technical support for evidence-based pollution control policy formulation in complex atmospheric environments.
{"title":"Machine learning-driven PMF modeling for accurate and objective source identification of VOCs","authors":"Qiaoli Wang , Xiaojie Ou , Chengzhi Wu , Shengdong Yao , Shenlin Huang , Chengcheng Zhu , Ziyi Liao , Kanghui Wang , Shihan Zhang , Jianmeng Chen","doi":"10.1016/j.apr.2025.102721","DOIUrl":"10.1016/j.apr.2025.102721","url":null,"abstract":"<div><div>Accurate source apportionment of volatile organic compounds (VOCs) serves as a critical foundation for developing targeted control strategies against complex air pollution. While the Positive Matrix Factorization (PMF) model has been extensively employed in pollution source identification, it exhibits inherent limitations when addressing long-term complex pollution scenarios, particularly manifested through slow convergence rates and unavoidable subjective biases during factor resolution. To address these technical challenges, this study innovatively proposes a machine learning-integrated PMF framework (<em>k</em>R-PMF), achieving effective synergy between machine learning algorithms and receptor modeling. The developed <em>k</em>R-PMF model demonstrated superior performance and improved R<sup>2</sup> values (0.76–0.95) between simulated and monitored data, indicating enhanced simulation stability. Source apportionment results revealed differential contributions from six major emission categories under PMF and <em>k</em>R-PMF frameworks: biogenic emission sources (8.6 % vs 13.7 %), solvent use (16.5 % vs 14.2 %), industrial emissions (13.6 % vs 10.8 %), gasoline volatilization (14.4 % vs 10.9 %), vehicle emissions (11.8 % vs 7.2 %), and combustion emissions (13.3 % vs 11.1 %). Notably, <em>k</em>R-PMF successfully resolved the secondary source formation process that accounts for 12.5 % of the total volatile organic compounds, a key component that is often obscured in traditional analyses. This methodological breakthrough establishes a robust framework for high-precision VOC source characterization, providing essential technical support for evidence-based pollution control policy formulation in complex atmospheric environments.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102721"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102729
Kedi Zhang , Yang Yang , Xiaoyu Ni , Qing Ma , Ye Yang , Ronghao Cai , Na Li
Maize production makes a significant contribution to global food security, yet it carries a high risk of ammonia (NH3) emissions. Field NH3 emissions consist of soil and canopy NH3 emissions. Among these, canopy NH3 emissions refer to NH3 losses from crop canopies, which are closely related to nitrogen (N) recovery efficiency. Many farmers prefer one-time N application to save labor, but little is known about the effects of one-time N application on soil and canopy NH3 emissions. This study aimed to investigate the effects of one-time N application on soil and canopy NH3 emissions in maize fields, thereby providing essential information for optimizing N management and reducing NH3 emissions. We conducted a two-year field experiment (2021 and 2022) with three treatments, i.e., control, one-time N application, and split N application. Soil and canopy NH3 emissions accounted for 78.2%–82.8% and 17.2%–21.8% of field NH3 emissions, respectively. Compared with split N application, one-time N application increased canopy NH3 emissions by 6.7%–14.3%, soil NH3 emissions by 4.3%–5.7%, field NH3 emissions by 4.7%–7.3%, and yield-scaled NH3 emissions by 11.4%–11.7%; while it reduced grain yield by 3.6%–6.2%, plant N uptake by 5.4%–8.0%, and N recovery efficiency by 10.2%–13.9%. Soil and canopy NH3 emissions in one-time N application treatment were driven by the higher soil NH4+ concentration, lower soil volumetric water content, and greater leaf apoplast NH4+ concentration and leaf area. These findings deepen our understanding of soil and canopy NH3 emissions and provide new insights into N management and NH3 emission reduction in maize production.
{"title":"One-time nitrogen application increased soil and canopy ammonia emissions in maize fields","authors":"Kedi Zhang , Yang Yang , Xiaoyu Ni , Qing Ma , Ye Yang , Ronghao Cai , Na Li","doi":"10.1016/j.apr.2025.102729","DOIUrl":"10.1016/j.apr.2025.102729","url":null,"abstract":"<div><div>Maize production makes a significant contribution to global food security, yet it carries a high risk of ammonia (NH<sub>3</sub>) emissions. Field NH<sub>3</sub> emissions consist of soil and canopy NH<sub>3</sub> emissions. Among these, canopy NH<sub>3</sub> emissions refer to NH<sub>3</sub> losses from crop canopies, which are closely related to nitrogen (N) recovery efficiency. Many farmers prefer one-time N application to save labor, but little is known about the effects of one-time N application on soil and canopy NH<sub>3</sub> emissions. This study aimed to investigate the effects of one-time N application on soil and canopy NH<sub>3</sub> emissions in maize fields, thereby providing essential information for optimizing N management and reducing NH<sub>3</sub> emissions. We conducted a two-year field experiment (2021 and 2022) with three treatments, i.e., control, one-time N application, and split N application. Soil and canopy NH<sub>3</sub> emissions accounted for 78.2%–82.8% and 17.2%–21.8% of field NH<sub>3</sub> emissions, respectively. Compared with split N application, one-time N application increased canopy NH<sub>3</sub> emissions by 6.7%–14.3%, soil NH<sub>3</sub> emissions by 4.3%–5.7%, field NH<sub>3</sub> emissions by 4.7%–7.3%, and yield-scaled NH<sub>3</sub> emissions by 11.4%–11.7%; while it reduced grain yield by 3.6%–6.2%, plant N uptake by 5.4%–8.0%, and N recovery efficiency by 10.2%–13.9%. Soil and canopy NH<sub>3</sub> emissions in one-time N application treatment were driven by the higher soil NH<sub>4</sub><sup>+</sup> concentration, lower soil volumetric water content, and greater leaf apoplast NH<sub>4</sub><sup>+</sup> concentration and leaf area. These findings deepen our understanding of soil and canopy NH<sub>3</sub> emissions and provide new insights into N management and NH<sub>3</sub> emission reduction in maize production.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102729"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102733
Neelam Baghel, Anita Lakhani, Aparna Satsangi, K. Maharaj Kumari
Volatile Organic Compounds (VOCs) are critical contributors to tropospheric ozone (O3) and Secondary Organic Aerosol (SOA) formation. Real time measurements of ambient VOCs were made at suburban site of Agra from August 2021 to July 2022 to assess their temporal and seasonal variation, possible sources, contribution to ozone formation potential (OFP) and secondary organic aerosol formation potential (SOAFP). VOCs reactivity with hydroxy (OH) radical was assessed to determine their photochemical loss. The average concentration of VOCs was 152.0 ± 75.2 μg/m3, of which, aromatic hydrocarbons were the most abundant compounds followed by alkanes, alkenes and aldehydes with higher levels in winter whereas isoprene and aldehydes had higher levels in summer period. OFP based on initial photochemical values of VOCs was found to be 84.4 % higher than measured value. SOAFP value was highest for the aromatic hydrocarbons which contributed more than 90 %. Positive matrix factorization (PMF) results showed emissions from compressed natural gas (CNG) vehicles and liquefied petroleum gas (LPG) from nearby residential areas were the major contributors (40.9 %) of VOCs. Carcinogenic effect of VOCs for adults and children was estimated by Integrated Lifetime Cancer Risk (ILCR) and non-carcinogenic effect by Hazard Quotient (HQ). ILCR and HQ values for benzene were higher for adults and children. The results of this study may be beneficial for predicting emission sources of VOCs in the region and for applying possible control measures to reduce ground-level O3.
{"title":"Real time monitoring of VOCs and their OH reactivity: Links with OFP and SOA formation at suburban site of Indo-Gangetic plain","authors":"Neelam Baghel, Anita Lakhani, Aparna Satsangi, K. Maharaj Kumari","doi":"10.1016/j.apr.2025.102733","DOIUrl":"10.1016/j.apr.2025.102733","url":null,"abstract":"<div><div>Volatile Organic Compounds (VOCs) are critical contributors to tropospheric ozone (O<sub>3</sub>) and Secondary Organic Aerosol (SOA) formation. Real time measurements of ambient VOCs were made at suburban site of Agra from August 2021 to July 2022 to assess their temporal and seasonal variation, possible sources, contribution to ozone formation potential (OFP) and secondary organic aerosol formation potential (SOA<sub>F</sub>P). VOCs reactivity with hydroxy (OH) radical was assessed to determine their photochemical loss. The average concentration of VOCs was 152.0 ± 75.2 μg/m<sup>3</sup>, of which, aromatic hydrocarbons were the most abundant compounds followed by alkanes, alkenes and aldehydes with higher levels in winter whereas isoprene and aldehydes had higher levels in summer period. OFP based on initial photochemical values of VOCs was found to be 84.4 % higher than measured value. SOA<sub>F</sub>P value was highest for the aromatic hydrocarbons which contributed more than 90 %. Positive matrix factorization (PMF) results showed emissions from compressed natural gas (CNG) vehicles and liquefied petroleum gas (LPG) from nearby residential areas were the major contributors (40.9 %) of VOCs. Carcinogenic effect of VOCs for adults and children was estimated by Integrated Lifetime Cancer Risk (ILCR) and non-carcinogenic effect by Hazard Quotient (HQ). ILCR and HQ values for benzene were higher for adults and children. The results of this study may be beneficial for predicting emission sources of VOCs in the region and for applying possible control measures to reduce ground-level O<sub>3</sub>.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102733"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102702
Mohammad Amin Javadi , Maryam Zare Shahne , Zahra Amiri
Nitrogen dioxide (NO2), a significant pollutant from human activities, poses immediate and long-term health risks. In our research, we utilized six distinct Machine Learning (ML) techniques including Multiple Linear Regression (MLR), Decision Tree Regression (DTR), Random Forest Regression (RFR), K-Nearest Neighbor Regression (KNNR), Support Vector Regression (SVR) and XGBoost for Regression, along with ground-based and satellite-derived data, to forecast NO2 levels at the Mandir Marg air quality control station in Delhi, India. This research also seeks to evaluate the efficiency of Sentinel-5P products using the Google Earth Engine (GEE) to monitor NO2 levels in Delhi from 2020 to 2021 and impacts of COVID-19. Our methodology involved inputting satellite imagery into GEE platform, which identified areas affected by pollutants hourly, daily, and monthly. We obtained NO2 concentrations from Sentinel-5P by employing JavaScript coding within GEE. The model used in this study, exclusively utilizing Sentinel-5P products, underwent evaluation and testing with ground-based data collected from the Central Control Room for Air Quality Management (CPCB). We also discussed the probable reasons for fluctuations in the values of NO2 in the study period, which were adjusted to previous research. Of the ML techniques employed, the RFR was the most accurate method in predicting NO2 concentrations. The methodology can describe the spatial and temporal fluctuations in NO2 concentrations, achieving the minimum root mean square error and the maximum R-squared. The findings of this research indicate that the integration of Sentinel-5P data with automated platforms like GEE aligns well with actual and predicted ground-based data.
{"title":"Integration of Sentinel-5P satellite data and machine learning for spatiotemporal prediction of NO2 in Delhi: Impacts of COVID-19 lockdown","authors":"Mohammad Amin Javadi , Maryam Zare Shahne , Zahra Amiri","doi":"10.1016/j.apr.2025.102702","DOIUrl":"10.1016/j.apr.2025.102702","url":null,"abstract":"<div><div>Nitrogen dioxide (NO<sub>2</sub>), a significant pollutant from human activities, poses immediate and long-term health risks. In our research, we utilized six distinct Machine Learning (ML) techniques including Multiple Linear Regression (MLR), Decision Tree Regression (DTR), Random Forest Regression (RFR), K-Nearest Neighbor Regression (KNNR), Support Vector Regression (SVR) and XGBoost for Regression, along with ground-based and satellite-derived data, to forecast NO<sub>2</sub> levels at the Mandir Marg air quality control station in Delhi, India. This research also seeks to evaluate the efficiency of Sentinel-5P products using the Google Earth Engine (GEE) to monitor NO<sub>2</sub> levels in Delhi from 2020 to 2021 and impacts of COVID-19. Our methodology involved inputting satellite imagery into GEE platform, which identified areas affected by pollutants hourly, daily, and monthly. We obtained NO<sub>2</sub> concentrations from Sentinel-5P by employing JavaScript coding within GEE. The model used in this study, exclusively utilizing Sentinel-5P products, underwent evaluation and testing with ground-based data collected from the Central Control Room for Air Quality Management (CPCB). We also discussed the probable reasons for fluctuations in the values of NO<sub>2</sub> in the study period, which were adjusted to previous research. Of the ML techniques employed, the RFR was the most accurate method in predicting NO<sub>2</sub> concentrations. The methodology can describe the spatial and temporal fluctuations in NO<sub>2</sub> concentrations, achieving the minimum root mean square error and the maximum R-squared. The findings of this research indicate that the integration of Sentinel-5P data with automated platforms like GEE aligns well with actual and predicted ground-based data.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102702"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102719
Abhishek Penchala, Aditya Kumar Patra
Understanding the emission and transport of mining-generated particulate matter (PM) is crucial for assessing its impact on miners and adjoining communities. This study evaluates the distribution of PM1, PM2.5, and PM10 mass concentrations and their interactions with key meteorological factors in vertical and horizontal directions at a large-scale surface coal mine. A low-cost sensor coupled with an unmanned aerial vehicle (UAV) was deployed to conduct a series of 32 flights (22 vertical and 10 horizontal). The vertical profiles from the pit bottom to the pit top (130 m vertical depth) have shown a nonuniform decrease in PM concentrations, with a higher percentage of decrease observed for PM10 (47 %) compared to PM2.5 (15 %) and PM1 (13 %). However, no consistent increasing or decreasing trend was observed in PM concentration in the horizontal flights up to a distance of 150 m from the pit boundary. Assessment of diurnal variations indicated higher PM levels during evening due to particle accumulation from continuous mining operations combined with a lower atmospheric boundary layer height. Airborne UAV measurements within and around the mine have shown dominance of PM2.5 and PM1 compared to their respective proportions at the ground. In addition to emissions from in-pit mining operations, microscopic analysis of PM revealed that mine workers and adjoining communities are exposed to PM2.5 originating from the heating of low-quality coal deposited along the mine benches. This critical observation highlights the contribution of PM from surface coal mining and the exposure risk faced by miners and surrounding communities.
{"title":"Assessment of vertical and horizontal distribution of respirable particulate matter in and around a surface coal mine","authors":"Abhishek Penchala, Aditya Kumar Patra","doi":"10.1016/j.apr.2025.102719","DOIUrl":"10.1016/j.apr.2025.102719","url":null,"abstract":"<div><div>Understanding the emission and transport of mining-generated particulate matter (PM) is crucial for assessing its impact on miners and adjoining communities. This study evaluates the distribution of PM<sub>1</sub>, PM<sub>2.5</sub>, and PM<sub>10</sub> mass concentrations and their interactions with key meteorological factors in vertical and horizontal directions at a large-scale surface coal mine. A low-cost sensor coupled with an unmanned aerial vehicle (UAV) was deployed to conduct a series of 32 flights (22 vertical and 10 horizontal). The vertical profiles from the pit bottom to the pit top (130 m vertical depth) have shown a nonuniform decrease in PM concentrations, with a higher percentage of decrease observed for PM<sub>10</sub> (47 %) compared to PM<sub>2.5</sub> (15 %) and PM<sub>1</sub> (13 %). However, no consistent increasing or decreasing trend was observed in PM concentration in the horizontal flights up to a distance of 150 m from the pit boundary. Assessment of diurnal variations indicated higher PM levels during evening due to particle accumulation from continuous mining operations combined with a lower atmospheric boundary layer height. Airborne UAV measurements within and around the mine have shown dominance of PM<sub>2.5</sub> and PM<sub>1</sub> compared to their respective proportions at the ground. In addition to emissions from in-pit mining operations, microscopic analysis of PM revealed that mine workers and adjoining communities are exposed to PM<sub>2.5</sub> originating from the heating of low-quality coal deposited along the mine benches. This critical observation highlights the contribution of PM from surface coal mining and the exposure risk faced by miners and surrounding communities.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102719"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102725
Haobo Cao , Weibin Ma , Jie Du , Weiyang Xiao , Lei He , Dingyong Wang
Long-term observations of atmospheric mercury concentrations and statistical methods based on observations are important tools for quantifying the impacts of anthropogenic and natural disturbances on the global atmospheric mercury pool. Jiuzhaigou is located in the transitional zone between the Qinghai-Tibet Plateau and the Sichuan Basin, at an altitude of 2000–3500 m in the subalpine zone. Investigating the dynamics of atmospheric mercury in Jiuzhaigou can provide a basis for in-depth analysis of mercury transport characteristics on the Qinghai-Tibet Plateau and global cycling processes. In this study, continuous monitoring of gaseous elemental mercury (GEM) in Jiuzhaigou was conducted for two years from 2021 to 2023 using a high time-resolution automatic mercury analyzer. The results showed that the average concentration of GEM in Jiuzhaigou was 1.25 ± 0.41 ng/m3, at the global background concentration of atmospheric mercury. However, the GEM concentration in Jiuzhaigou exhibited significant temporal variations, with higher concentrations in winter and lower concentrations in summer, and a daily variation pattern of higher concentrations at night and lower concentrations during the day. The effects of meteorological factors on GEM concentration were quantified using Generalized Additive Models (GAMs), indicating that relative humidity had a significant impact on GEM, and the inter-annual differences in GEM may also be influenced by atmospheric pressure and wind speed. The Potential Source Contribution Function (PSCF) model based on backward trajectories analyzed the variations of potential source areas of GEM in the atmosphere in Jiuzhaigou during spring and winter, while local air masses near the boundary layer height dominated the input in summer and autumn.
{"title":"Atmospheric mercury dynamics in subalpine regions of the Tibetan Plateau and its influencing factors","authors":"Haobo Cao , Weibin Ma , Jie Du , Weiyang Xiao , Lei He , Dingyong Wang","doi":"10.1016/j.apr.2025.102725","DOIUrl":"10.1016/j.apr.2025.102725","url":null,"abstract":"<div><div>Long-term observations of atmospheric mercury concentrations and statistical methods based on observations are important tools for quantifying the impacts of anthropogenic and natural disturbances on the global atmospheric mercury pool. Jiuzhaigou is located in the transitional zone between the Qinghai-Tibet Plateau and the Sichuan Basin, at an altitude of 2000–3500 m in the subalpine zone. Investigating the dynamics of atmospheric mercury in Jiuzhaigou can provide a basis for in-depth analysis of mercury transport characteristics on the Qinghai-Tibet Plateau and global cycling processes. In this study, continuous monitoring of gaseous elemental mercury (GEM) in Jiuzhaigou was conducted for two years from 2021 to 2023 using a high time-resolution automatic mercury analyzer. The results showed that the average concentration of GEM in Jiuzhaigou was 1.25 ± 0.41 ng/m<sup>3</sup>, at the global background concentration of atmospheric mercury. However, the GEM concentration in Jiuzhaigou exhibited significant temporal variations, with higher concentrations in winter and lower concentrations in summer, and a daily variation pattern of higher concentrations at night and lower concentrations during the day. The effects of meteorological factors on GEM concentration were quantified using Generalized Additive Models (GAMs), indicating that relative humidity had a significant impact on GEM, and the inter-annual differences in GEM may also be influenced by atmospheric pressure and wind speed. The Potential Source Contribution Function (PSCF) model based on backward trajectories analyzed the variations of potential source areas of GEM in the atmosphere in Jiuzhaigou during spring and winter, while local air masses near the boundary layer height dominated the input in summer and autumn.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102725"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.apr.2025.102699
Shan-Yi Shen , Ru-Yun Chang , Yun-Wei Liao , Wei-Chun Chou , Yu-Hao Lin , Ping-Yu Liu
Providing a high-resolution spatiotemporal concentration of traffic pollutants can support more effective traffic pollution control. The concentration of on-road carbon monoxide (CO) originating from vehicle engine combustion usually has a high positive correlation with traffic volume. Hence, the innovative development of this On-road Vehicle Emission Estimation Model (OVEEM), built with a bottom-up framework, aimed to deliver the high-resolution spatiotemporal data on CO distribution. Based on the technology of web crawler and deep learning, OVEEM not only collect public vehicle detectors and public traffic Surveillance Videos (SVs) but also estimate both the traffic volume and the average velocities of vehicles. The hourly CO emissions can be estimated by multiplying the hourly traffic volume by the CO Emission Factor (EF) corresponding to the average vehicle speed. The emissions from total vehicles were input into the AERMOD dispersion model to estimate the spatiotemporal distributions of CO concentrations. Finally, GIS was employed to visualize the high-resolution spatiotemporal distributions of CO concentrations.
The result indicates that both the observed and the estimated concentrations of CO followed similar trends over the period of 24 h, with a reasonable mean absolute percent error between them. These findings validated the proposed OVEEM through a comparison between the observed and the estimated CO concentrations. Also, scooters and sedans were found to be the main types of vehicles contributing to elevated CO concentrations. In the future, to estimate the distribution of other pollutants, more appropriate SVs should be obtained.
{"title":"A proposed framework based on traffic volume and the design of a bottom-up methodology to estimate the CO emission of on-road vehicles","authors":"Shan-Yi Shen , Ru-Yun Chang , Yun-Wei Liao , Wei-Chun Chou , Yu-Hao Lin , Ping-Yu Liu","doi":"10.1016/j.apr.2025.102699","DOIUrl":"10.1016/j.apr.2025.102699","url":null,"abstract":"<div><div>Providing a high-resolution spatiotemporal concentration of traffic pollutants can support more effective traffic pollution control. The concentration of on-road carbon monoxide (CO) originating from vehicle engine combustion usually has a high positive correlation with traffic volume. Hence, the innovative development of this On-road Vehicle Emission Estimation Model (OVEEM), built with a bottom-up framework, aimed to deliver the high-resolution spatiotemporal data on CO distribution. Based on the technology of web crawler and deep learning, OVEEM not only collect public vehicle detectors and public traffic Surveillance Videos (SVs) but also estimate both the traffic volume and the average velocities of vehicles. The hourly CO emissions can be estimated by multiplying the hourly traffic volume by the CO Emission Factor (EF) corresponding to the average vehicle speed. The emissions from total vehicles were input into the AERMOD dispersion model to estimate the spatiotemporal distributions of CO concentrations. Finally, GIS was employed to visualize the high-resolution spatiotemporal distributions of CO concentrations.</div><div>The result indicates that both the observed and the estimated concentrations of CO followed similar trends over the period of 24 h, with a reasonable mean absolute percent error between them. These findings validated the proposed OVEEM through a comparison between the observed and the estimated CO concentrations. Also, scooters and sedans were found to be the main types of vehicles contributing to elevated CO concentrations. In the future, to estimate the distribution of other pollutants, more appropriate SVs should be obtained.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"17 1","pages":"Article 102699"},"PeriodicalIF":3.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145973624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号