Atmospheric Pollution Research最新文献

{"title":"Association between short-term exposure to PM2.5 and mortality in adults with dyslipidemia and asthma: a time-stratified case-crossover study in South Korea","authors":"Min Chae Kim ,&nbsp;Hyeokjong Lee ,&nbsp;Sun Jae Park ,&nbsp;Jaewon Kim ,&nbsp;Hye Jun Kim ,&nbsp;Jihun Song ,&nbsp;Sangwoo Park ,&nbsp;Hyun-Young Shin ,&nbsp;Hong Yun Jung ,&nbsp;Seung Ju Choi ,&nbsp;Youn Jae Lee ,&nbsp;Hyoung Gil Yoon ,&nbsp;Seong Hyok Kim ,&nbsp;Sang Min Park","doi":"10.1016/j.apr.2025.102658","DOIUrl":"10.1016/j.apr.2025.102658","url":null,"abstract":"<div><div>The harmful effects of particulate matter (PM) on health are well-documented, and recent studies have increasingly focused on its impact on vulnerable populations. While PM exposure is known to affect both lipid levels and asthma exacerbation, its impact on mortality risk in individuals with coexisting dyslipidemia and asthma has not been well studied, despite evidence linking these two conditions.</div><div>Using the National Health Insurance Service (NHIS) database, we identified 51,833 patients with dyslipidemia and newly diagnosed asthma who died between January 1, 2015, and December 31, 2021. The study period spanned January 1, 2009, to December 31, 2021, for baseline data collection. The association between short-term PM_2.5 exposure and mortality was analyzed with conditional logistic regression using a time-stratified case-crossover design.</div><div>Short-term PM_2.5 exposure was significantly associated with increased mortality risk among individuals with dyslipidemia and newly diagnosed asthma. Across various lag periods and models, the odds ratio per 10 μg/m³ increase in PM_2.5, with consistent trends of statistical significance (p-trend &lt;0.05).</div><div>This study is the first to demonstrate that increased short-term exposure to PM_2.5 elevates mortality risk in patients with comorbid dyslipidemia and asthma. Efforts to reduce PM_2.5 emissions and establish systematic PM concentration alert systems are expected to improve health outcomes for these patients.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102658"},"PeriodicalIF":3.9,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679699","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}

{"title":"Pollution characteristic and temporal trend of PM2.5-bound heavy metals in a megacity of Northeast China: Implication for the effect of air pollution control policy","authors":"Jin-Wen He ,&nbsp;Li-Yan Liu ,&nbsp;Chang-Yu Weng ,&nbsp;De-Qi Wang ,&nbsp;Shi-Ming Jia ,&nbsp;Wan-Li Ma","doi":"10.1016/j.apr.2025.102660","DOIUrl":"10.1016/j.apr.2025.102660","url":null,"abstract":"<div><div>PM_2.5-bound heavy metals present significant health risks especially in cold regions due to the enhanced coal combustion for space heating. This study conducted a comprehensive one-year observation of PM_2.5-bound heavy metals in Harbin City, a megacity in Northeast China, from August 2021 to July 2022. The findings revealed that zinc (Zn) and chromium (Cr) were the dominant heavy metals, with concentrations of 47 ng/m³ and 41 ng/m³, respectively. Notably, the levels of Cr(VI) exceeded both the WHO guidelines and the Chinese Ambient Air Quality Standards. Except for Cr, the concentrations of other heavy metals were significantly higher during the heating period than the non-heating period. Based on source apportionment, traffic emission and coal combustion were recognized as the primary sources in both periods, while coal combustion was the dominant source in the heating period. Health risk assessments revealed non-carcinogenic risk from arsenic (As) for infants and toddlers, while carcinogenic risk for most elements surpassed the threshold of 10⁻⁶. Nickel (Ni) and Cr(VI) emerged as the principal carcinogenic elements during the heating period and non-heating period, respectively. A comparative assessment demonstrated a substantial decline in the concentrations of targeted heavy metals, with an overall reduction of 75.9% compared to 2013. Importantly, the heating period accounted for 65.6% of this reduction, suggesting the effectiveness of air pollution control policies, especially those aimed at reducing coal combustion emissions. Therefore, the control policy addressing the specific sources of heavy metals is essential.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102660"},"PeriodicalIF":3.9,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679698","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}

{"title":"Uncertainty assessment based on data decomposition and Boruta-driven extreme gradient boosting to predict spatiotemporal urban air dust heavy metal index","authors":"Akram Seifi ,&nbsp;Somayeh Soltani-Gerdefaramarzi ,&nbsp;Mumtaz Ali","doi":"10.1016/j.apr.2025.102654","DOIUrl":"10.1016/j.apr.2025.102654","url":null,"abstract":"<div><div>Accurate prediction of urban air dust pollutants is essential for public health and environmental management. Achieving reliable predictions of the air pollution due to heavy metals existence in these areas is extremely important. This study for the first time develop an ensemble approach based on multivariate variational model decomposition (MVMD) and extreme gradient boosting (XGBoost) integrated with Bayesian optimizer of Optuna and different feature selection techniques to predict the spatiotemporal distribution of pollution load index (PLI) in Yazd urban area, Iran. For comparison, gated recurrent unit (GRU) network, adaptives neuro-fuzzy-inference system (ANFIS), and multilayer perceptron (MLP) models were are develpoed. Variables including meteorological data, heavy metals concentration of roof dust, and distance to pollution sources were gathered. The seasonal data of variables were analyzed using Boruta feature selection approach (BFSA), SHapley additive explanations (SHAP), and Wavelet methods to identify valuable and easily accessible variables to predict PLI index. The results confirmed that the BFSA has high capability for selecting the most important features over SHAP, and wavelet techniques, that provides cost-effective input vector of Max WD, Min RH, Cd, and Zn with readily available variables. Morover, the XGBoost model shows high prediction accuracy for PLI in terms of R² = 0.90, RMSE = 0.08, and MAE = 0.06. Furthermore, by stationarity test of multivariate variational mode decomposition (MVMD) method applied to all input variables, the Max WD and Min RH were decompossed into three intrinsic mode functions (IMFs). These IMFs along with Cd and Zn were used as input vector in the XGBoost to create the final model for predicting temporal uncertainty and generate seasonal urban spatiotemporal maps. The evaluation of uncertainties demonstrated that the MVMD-XGBoost effectively captured 83.33 %, 96.67 %, 63.33 %, and 68.97 % of observed data within the 95 % confidence interval in spring, summer, autumn, and winter seasons, respectively. Findings from this study allow decision-makers to reduce air pollution monitoring costs and enhance control measures by leveraging readily available variables.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102654"},"PeriodicalIF":3.9,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144714512","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}

{"title":"Assessment of IoT low-cost sensor networks for long-term outdoor and indoor air quality monitoring: a case study in Dublin","authors":"Jiayao Chen ,&nbsp;Óscar González ,&nbsp;David O'Connor ,&nbsp;Lindsay Tallon ,&nbsp;Francesco Pilla","doi":"10.1016/j.apr.2025.102651","DOIUrl":"10.1016/j.apr.2025.102651","url":null,"abstract":"<div><div>This study provides a framework for Internet of Things based low-cost sensors (LCS) network implementation, using office environments in Dublin, Ireland, as a case study for long-term indoor air quality (IAQ) monitoring. It covers options and key decisions related to sensor technology, reporting systems and data management. Environmental and indoor data were collected from 1 June 2023 to 20 June 2024, using Smart Citizen Kit 2.1 and PurpleAir devices, and data retrieved from cloud-based data platforms for analysis. The standard deviation and coefficient of variation were calculated to evaluate intra-sensor precision. To improve data quality of LCS various correction models were tested, considering the impact of temperature and relative humidity. A multilinear model with additive relative humidity, using the piecewise regression, provided better performance (R² &gt; 0.7, RMSE &lt;5 μg/m³) and accuracy (&gt;0.88) for 24-h fine particulate matter (PM_2.5) and inhalable particulate matter (PM₁₀). This study bridges the data gap by incorporating multi-brand LCS network for further application in outdoor supplementary and IAQ reporting. The results showed corrected indoor PM_2.5 data in offices complies with WHO air quality guidelines, and carbon dioxide (CO₂) levels in naturally ventilated conditions remained below 800 ppm. Additionally, diurnal patterns reveal elevated levels of CO₂ and total volatile organic compounds during core office hours, while the contrasting patterns for PM_2.5 suggest outdoor infiltration as the dominant source. This study demonstrates the potential of data-driven techniques for real-time IAQ monitoring and reporting, providing valuable insights to promote healthier IAQ for occupants.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102651"},"PeriodicalIF":3.9,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144679704","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}

{"title":"Source apportionment of primary and secondary fine particulate matter in Eskisehir/Türkiye using conventional and dispersion-normalized positive matrix factorization","authors":"Akif Ari ,&nbsp;Pelin Ertürk Ari ,&nbsp;Eftade O. Gaga","doi":"10.1016/j.apr.2025.102653","DOIUrl":"10.1016/j.apr.2025.102653","url":null,"abstract":"<div><div>In this study, the organic and inorganic components of atmospheric fine particulate matter (PM_2.5) were monitored daily in Eskisehir City/Türkiye, for one year, and the contributions of primary and secondary sources to the PM mass were investigated. A total of 94 components were characterized in PM_2.5 samples, including 5 anions, 46 trace elements, Organic and Elemental Carbon (OC &amp; EC), 16 Polycyclic Aromatic Hydrocarbons (PAHs), 26 n-alkanes, levoglucosan, and 8 carboxylic acids to obtain a more holistic mass closure. In addition, the principal sources of PM_2.5 were apportioned by the Conventional and Dispersion-Normalized Positive Matrix Factorization (C-PMF and DN-PMF) to assess dispersion/dilution characteristics of local meteorological conditions on source contributions of PM. The main sources of PM were classified into 8 factors. A significant seasonal variation was observed in combustion-related PM_2.5 constituents, which increased in the winter, while the contribution of Secondary Organic Carbon (SOC) enhanced during the summer period. In addition, a visual effect of seasonal atmospheric dilution/dispersion conditions on measured pollutant levels was observed as a function of Ventilation Coefficients (VC). The mass percent of SOC in PM_2.5 varied between 2.3 % and 13.0 %, and the annual mean contribution was over 7.5 %.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102653"},"PeriodicalIF":3.9,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144633947","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}

{"title":"The input-optimized and supplemented MEGAN improves biogenic volatile organic compounds estimation accuracy in China","authors":"Bin Cai ,&nbsp;Haomiao Cheng ,&nbsp;Tianfang Kang ,&nbsp;Yu Wang ,&nbsp;Wentao Han ,&nbsp;Wei Wang","doi":"10.1016/j.apr.2025.102650","DOIUrl":"10.1016/j.apr.2025.102650","url":null,"abstract":"<div><div>To reduce the uncertainty in biogenic volatile organic compound (BVOC) emission estimates by MEGAN model, this study introduces targeted optimizations in the input conditions driven by plant functional types (PFTs), emission factors (EFs), and leaf area index (LAI), and supplements the missing BVOC emissions from urban green spaces (U-BVOC emissions). To verify the optimization effect of the input-improved MEGAN, the study conducted a comparative validation of the estimated BVOC emission levels in representative regions of northern and southern China using formaldehyde vertical column densities and accumulation-based estimates through stock volume conversion. The results indicate that the input-improved MEGAN can provide more reasonable BVOC estimates with a more notable improvement in the southern region, and the estimated isoprene emissions show better spatiotemporal correlation with formaldehyde vertical column densities. The optimization of PFT + EF and LAI input conditions both effectively improve the model's estimation accuracy. Although the U-BVOC emissions are significantly lower than the BVOC emissions from non-urban environment, the newly added U-BVOC emission inventory is expected to help improve the performance of existing air quality models in simulating O₃ and particulate matter pollution in urban areas.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102650"},"PeriodicalIF":3.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144703477","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}

{"title":"Effects of altitude on light gasoline vehicles: fuel consumption and air pollution","authors":"Yanxu Ren,&nbsp;Wenhan Yu,&nbsp;Lijun Hao,&nbsp;Yunshan Ge","doi":"10.1016/j.apr.2025.102652","DOIUrl":"10.1016/j.apr.2025.102652","url":null,"abstract":"<div><div>Driven by the global carbon neutrality goal and emission regulations, gasoline vehicles' fuel consumption and emission characteristics in the plateau environment have received extensive attention. In this paper, the running resistance of light gasoline vehicles in a high-altitude environment laboratory is corrected by coasting resistance tests to ensure the accuracy of laboratory tests. Finally, a constant speed fuel consumption test and a New European Driving Cycle emission test were carried out in the high-altitude environment laboratory. The results of fuel consumption characteristics show that at low-to-medium speeds (below 5,000m altitude), fuel efficiency generally improved with increasing altitude. For the high-speed range, air resistance becomes more prominent below 5000m, which reduces overall fuel consumption. However, once the altitude approaches 5000m, all speed ranges show an upward trend in fuel consumption. Fuel consumption at high speeds enters an upward trend earlier (at relatively low altitudes). The results of emission characteristics show that the emission of CO and THC increases significantly with the increase of altitude due to combustion deterioration. The contribution of the cold start-up phase increased significantly (77.1 % CO and 66.3 % THC), which became the core factor of the increase in emissions. Due to the game between the combustion temperature and the efficiency of the three-way catalytic converter (TWC), NOx emission shows a trend of first decreasing and then increasing. Its instantaneous emission characteristics are not stable because of the gradual decline of TWC performance. In general, these results can be used to drive vehicle energy efficiency and emission reduction targets.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102652"},"PeriodicalIF":3.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144695222","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}

{"title":"Long-term evolution and synoptic meteorological modulation of PM2.5 and PM10 in Seoul","authors":"Wooseok Jang ,&nbsp;Simon Wang ,&nbsp;Hyun Cheol Kim ,&nbsp;Jee-Hoon Jeong ,&nbsp;Changhyun Yoo ,&nbsp;Jin-Young Choi ,&nbsp;Jin-Ho Yoon","doi":"10.1016/j.apr.2025.102649","DOIUrl":"10.1016/j.apr.2025.102649","url":null,"abstract":"<div><div>Addressing particulate matter (PM) pollution in megacities like Seoul is crucial for public health and environmental sustainability, necessitating a comprehensive understanding of its long-term evolution and meteorological drivers. This study investigated the key factors affecting long-term PM concentrations in Seoul, South Korea, from 2000 to 2021, with a focus on the winter (DJF) and spring (MAM) seasons. To address the gap in PM research caused by the shorter observation period of PM_2.5 compared with PM₁₀, we used an extended PM_2.5 dataset. This enabled a detailed analysis of PM_2.5 and its relationship with PM₁₀, which, despite some differences, generally displayed similar variability. Both PM_2.5 and PM₁₀ exhibited decreasing trends in winter as well as spring, although the rate of decline slowed in the last decade (2011–2020) compared with the earlier decade (2000–2010). Both seasons exhibited a strengthened interannual correlation between PM_2.5 and PM₁₀ in the last decade. Daily PM_2.5 and PM₁₀ levels generally fluctuated in a similar pattern in both seasons, which can be attributed to synoptic-scale meteorological systems, particularly migratory systems from Northwest China, which can remain stationary over Korea for several days, particularly in winter. This pattern continues into spring, albeit with a lower intensity. These findings provide valuable insights into PM_2.5 variability and its correlation with PM₁₀ over time, which may inform future PM_2.5 mitigation strategies.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102649"},"PeriodicalIF":3.9,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144703476","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}

{"title":"Characteristics and influencing factors of ambient ozone pollution in Hangzhou in the relative humidity range with high ozone levels","authors":"Yu Lu ,&nbsp;Lingdong Kong ,&nbsp;Jiandong Shen ,&nbsp;Beibei Liu ,&nbsp;Yixuan An ,&nbsp;Yuwen Wang ,&nbsp;Jie Tan ,&nbsp;Lin Wang","doi":"10.1016/j.apr.2025.102648","DOIUrl":"10.1016/j.apr.2025.102648","url":null,"abstract":"<div><div>With the rapid economic growth and urbanization, despite improvements in particulate matter (PM_2.5) pollution control, ozone (O₃) pollution has emerged as a pressing environmental issue in China. This study systematically investigates O₃ pollution dynamics in Hangzhou (2021–2023) in the relative humidity (RH) range with high ozone levels using observational data and generalized additive modeling (GAM). Key findings reveal distinct temporal patterns: diurnal O₃ peaks at 14:00 (lagging solar radiation by about 2 h), seasonal maxima in summer (157.58 μg m⁻³) driven by temperature-photochemistry coupling, and non-monotonic annual trends (116.21–123.77 μg m⁻³) despite NO_x decline, reflecting transitional chemical regimes. Alkenes (36.94 %) and oxygenated VOCs (OVOCs, 36.54 %) dominated O₃ formation potential, with acetaldehyde, ethylene, and 1-butene as top contributors, highlighting the significant contributions of industrial and vehicular emission sources to O₃ formation. GAM analysis identified temperature as the primary driver, exhibiting exponential O₃ enhancement above 30 °C. In particular, NO₂ and peroxyacetyl nitrate showed synergistic effects, suggesting their dual roles as both precursors and indicators of radical cycling efficiency under the RH range. The 40 %–60 % RH range optimizes hydroxyl radical production while minimizes aerosol hydration, establishing it as a critical threshold for photochemical O₃ generation. These findings emphasize the importance of considering specific RH ranges and precursor reactivity in formulating refined O₃ pollution control strategies, providing a new theoretical basis for O₃ pollution control in the Yangtze River Delta region.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102648"},"PeriodicalIF":3.9,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144597291","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}

{"title":"Improving PM2.5 simulations using LSTM: a study on spatiotemporal generalization","authors":"Xingtian Chen ,&nbsp;Yuhang Zhang ,&nbsp;Kai Cao ,&nbsp;Dongxing Li ,&nbsp;Qizhong Wu","doi":"10.1016/j.apr.2025.102647","DOIUrl":"10.1016/j.apr.2025.102647","url":null,"abstract":"<div><div>Deep learning models, particularly long short-term memory (LSTM) networks, have shown strong potential for bias correction in air quality simulations. Although LSTM-based models have demonstrated success in improving PM_2.5 simulations during specific periods such as winter months or heavy pollution events, their ability to generalize across varying temporal intervals and geographical locations remains underexplored. This study systematically investigates the spatiotemporal generalization capabilities of LSTM-based (LSTM_Local, LSTM_Regional, LSTM_Regional-idx, LSTM_Regional-sub) models across 12 monthly intervals and 34 monitoring sites in Beijing. Benchmark comparisons with alternative deep learning models (RNN, GRU, BPNN, CNN) demonstrate the superiority of LSTM-based model in improving predictive performance. Seasonal analysis reveals that the LSTM_Local and LSTM_Regional-sub model achieved modest gains during summer half-year (May to October) but achieved significant improvements during winter half-year (November to April), with average RMSE reductions of −5.89 % (November), −17.40 % (December), −6.37 % (January), 0.36 % (February), −5.87 % (March) and −1.57 % (April). Spatially, urban sites show moderate gains, but suburban sites exhibit greater improvement, with average RMSE of −5.14 % (Center), −2.26 % (South-East), −7.05 % (North-East), −8.39 % (South-West) and −12.37 % (North-West). These findings highlight the robust spatiotemporal generalization of LSTM-based models and support their applicability for real-time bias correction, long-term forecasting, and air quality dataset enhancement at fine spatial and temporal resolutions.</div></div>","PeriodicalId":8604,"journal":{"name":"Atmospheric Pollution Research","volume":"16 11","pages":"Article 102647"},"PeriodicalIF":3.9,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144686007","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}