Search Results (191)

This study analyzes air pollution time-series big data to assess stationarity, seasonal patterns, and the performance of machine learning models in forecasting PM2.5 concentrations. Fifty-two low-cost sensors (LCS) were deployed across Krakow city and its surroundings (Poland), collecting hourly air quality data and generating nearly 20,000 observations per month. The network captured both spatial and temporal variability. The Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test confirmed trend-based non-stationarity, which was addressed through differencing, revealing distinct daily and 12 h cycles linked to traffic and temperature variations. Additive seasonal decomposition exhibited time-inconsistent residuals, leading to the adoption of multiplicative decomposition, which better captured pollution outliers associated with agricultural burning. Machine learning models—Ridge Regression, XGBoost, and LSTM (Long Short-Term Memory) neural networks—were evaluated under high spatial and temporal variability (winter) and low variability (summer) conditions. Ridge Regression showed the best performance, achieving the highest $R^2$ (0.97 in winter, 0.93 in summer) and the lowest mean squared errors. XGBoost showed strong predictive capabilities but tended to overestimate moderate pollution events, while LSTM systematically underestimated PM2.5 levels in December. The residual analysis confirmed that Ridge Regression provided the most stable predictions, capturing extreme pollution episodes effectively, whereas XGBoost exhibited larger outliers. The study proved the potential of low-cost sensor networks and machine learning in urban air quality forecasting focused on rare smog episodes (RSEs). Full article

The occurrence of haze pollution significantly deteriorates air quality and threatens human health, yet persistent knowledge gaps in real-time source apportionment of fine particulate matter (PM_2.5) hinder sustained improvements in atmospheric pollution conditions. Thus, this study employed single-particle aerosol mass spectrometry (SPAMS) to investigate PM_2.5 sources and dynamics during winter haze episodes in Yinchuan, Northwest China. Results showed that the average PM_2.5 concentration was 57 μg·m⁻³, peaking at 218 μg·m⁻³. PM_2.5 was dominated by organic carbon (OC, 17.3%), mixed carbonaceous particles (ECOC, 17.0%), and elemental carbon (EC, 14.3%). The primary sources were coal combustion (26.4%), fugitive dust (25.8%), and vehicle emissions (19.1%). Residential coal burning dominated coal emissions (80.9%), highlighting inefficient decentralized heating. Source contributions showed distinct diurnal patterns: coal combustion peaked nocturnally (29.3% at 09:00) due to heating and inversions, fugitive dust rose at night (28.6% at 19:00) from construction and low winds, and vehicle emissions aligned with traffic (17.5% at 07:00). Haze episodes were driven by synergistic increases in local coal (+4.0%), dust (+2.7%), and vehicle (+2.1%) emissions, compounded by regional transport (10.1–36.7%) of aged particles from northwestern zones. Fugitive dust correlated with sulfur dioxide (SO₂) and ozone (O₃) (p < 0.01), suggesting roles as carriers and reactive interfaces. Findings confirm local emission dominance with spatiotemporal heterogeneity and regional transport influence. SPAMS effectively resolved short-term pollution dynamics, providing critical insights for targeted air quality management in arid regions. Full article

Air quality forecasts play a crucial role in informing the public about atmospheric pollutant levels that pose risks to human health and the environment. The accuracy of these forecasts strongly depends on the quality and resolution of the input data used in the modelling process. At HungaroMet, the Hungarian Meteorological Service, the CHIMERE chemical transport model is used to provide two-day air quality forecasts for the territory of Hungary. This study compares two configurations of the CHIMERE model: the current operational setup, which uses climatological averages from the LMDz-INCA database for boundary conditions, and a test configuration that incorporates real-time boundary conditions from the CAMS global forecast. The primary objective of this work was to assess how the use of real-time versus climatological boundary conditions affects modelled concentrations of key pollutants, including NO₂, O₃, PM₁₀, and PM_2.5. The model results were evaluated against observational data from the Hungarian Air Quality Monitoring Network using a range of statistical metrics. The results indicate that the use of real-time boundary conditions, particularly for aerosol-type pollutants, improves the accuracy of PM₁₀ forecasts. This improvement is most significant under meteorological conditions that favour the long-range transport of particulate matter, such as during Saharan dust or wildfire episodes. These findings highlight the importance of incorporating dynamic, up-to-date boundary data, especially for particulate matter forecasting—given the increasing frequency of transboundary dust events. Full article

Accurately identifying the sources of fine particulate matter (PM_2.5) pollution is crucial for pollution control and public health protection. Taking the PM_2.5 pollution event that occurred in Suzhou in June 2023 as a typical case, this study analyzed the characteristics and components of PM_2.5, and quantified the contributions of meteorological conditions, regional transport, and local emissions to the summertime PM_2.5 surge in a typical Yangtze River Delta (YRD) city. Chemical composition analysis highlighted a sharp increase in nitrate ions (NO₃⁻, contributing up to 49% during peak pollution), with calcium ion (Ca²⁺) and sulfate ion (SO₄²⁻) concentrations rising to 2 times and 7.5 times those of clean periods, respectively. Results from the random forest model demonstrated that emission sources (74%) dominated this pollution episode, significantly surpassing the meteorological contribution (26%). The Weather Research and Forecasting model combined with the Community Multiscale Air Quality model (WRF–CMAQ) further revealed that local emissions contributed the most to PM_2.5 concentrations in Suzhou (46.3%), while external transport primarily originated from upwind cities such as Shanghai and Jiaxing. The findings indicate synergistic effects from dust sources, industrial emissions, and mobile sources. Validation using electricity consumption and key enterprise emission data confirmed that intensive local industrial activities exacerbated PM_2.5 accumulation. Recommendations include strengthening regulations on local industrial and mobile source emissions, and enhancing regional joint prevention and control mechanisms to mitigate cross-boundary transport impacts. Full article

China’s accelerating industrialization has led to worsening air pollution, characterized by recurrent haze episodes. The accurate quantification of PM_2.5 distribution is crucial for air quality assessment and public health management. Although traditional prediction models can effectively identify PM_2.5 concentration fluctuations with moderate accuracy, their dependence relies heavily on extensive ground-based monitoring station data, limiting their applicability in areas with sparse monitoring coverage. To address this limitation, this study proposes a novel algorithm for high-precision PM_2.5 concentration prediction, termed VDMS (Vision Transformer with DLSTM Multi-Head Self-Attention and Self-supervision). Based on the traditional Vision Transformer (ViT) architecture, VDMS incorporates a Double-Layered Long Short-Term Memory (DLSTM) network and a Multi-Head Self-Attention mechanism to enhance the model’s capacity to capture temporal sequence features and global dependencies. These enhancements contribute to greater stability and robustness in feature representation, ultimately improving prediction performance. Cross-validation experimental results show that the VDMS model outperforms benchmark models in PM_2.5 concentration prediction tasks, achieving a coefficient of determination (R²) of 0.93, a root mean square error (RMSE) of 4.05 μg/m³, and a mean absolute error (MAE) of 3.23 μg/m³. Furthermore, experiments conducted in areas with sparse ground monitoring stations demonstrate that the model maintains high predictive accuracy, further validating its applicability and generalization capability in data-limited scenarios. Moreover, the VDMS model adopts a modular design, offering strong scalability that allows its architecture to be adjusted according to specific requirements. This adaptability renders it suitable for monitoring various atmospheric pollutants, providing essential technical support for precise environmental management and air quality forecasting. Full article

Since 1989, California pollution control efforts have caused annual PM_2.5 averages to decrease. Despite the decline in ambient air concentrations of PM_2.5, the San Joaquin Valley (SJV) of California continues to violate the federal standard for PM_2.5. This study evaluated PM_2.5 trends, diurnal and seasonal patterns, pollution sources, and air quality improvements from 2000 to 2022 in the SJV. Hourly and daily PM_2.5 data from CARB and EPA-certified monitors were analyzed using regression models, polar plots, and Air Quality Index (AQI) classification methods. Monthly PM_2.5 concentrations peaked in winter (November–January) and during commute periods, with higher levels observed on Fridays and Saturdays. In this study, the highest daily PM_2.5 levels observed in Fresno and Bakersfield occurred during the autumn, most likely due to agricultural activities and higher wind speeds, with daily values greater than 25 µgm⁻³ and 50 µgm⁻³, respectively. In contrast, in Clovis, the highest daily PM_2.5 concentrations occurred in the winter during episodes characterized by low wind speeds, with values greater than 22 µgm⁻³. While PM_2.5 has declined since 1999, progress has slowed significantly since 2010. However, all sites exceeded the new EPA standard of 9 µgm⁻³. Without substantial changes to emission sources, meeting federal standards will be difficult. Full article

This study presented a comprehensive physical–statistical analysis of atmospheric particulate matter (PM₁₀ and PM_2.5) and trace gases (SO₂ and O₃) over Faial Island in the Azores archipelago during 2024. We collected real-time data at the Espalhafatos rural background station, covering 35,137 observations per pollutant, with 15 min intervals. Descriptive statistics, probability distribution fitting (Normal, Lognormal, Weibull, Gamma), and correlation analyses were employed to characterize pollutant dynamics and identify extreme pollution episodes. The results revealed that PM_2.5 (fine particles) concentrations are best modeled by a Lognormal distribution, while PM₁₀ concentrations fit a Gamma distribution, highlighting the presence of heavy-tailed, positively skewed behavior in both cases. Seasonal and episodic variability was significant, with multiple Saharan dust transport events contributing to PM exceedances, particularly during winter and spring months. These events, confirmed by CAMS and SKIRON dust dispersion models, affected not only southern Europe but also the Northeast Atlantic, including the Azores region. Weak to moderate correlations were observed between PM concentrations and meteorological variables, indicating complex interactions influenced by atmospheric stability and long-range transport processes. Linear regression analyses between SO₂ and O₃, and between SO₂ and PM_2.5, showed statistically significant but low-explanatory relationships, suggesting that other meteorological and chemical factors play a dominant role. This result highlights the importance of developing air quality policies that address both local emissions and long-range transport phenomena. They support the implementation of early warning systems and health risk assessments based on probabilistic modeling of particulate matter concentrations, even in remote Atlantic locations such as the Azores. Full article

PM₁₀ samples were collected at an urban site of Zhuozhou, the southern gateway of Beijing, from 28 December 2021 to 29 January 2022, in order to explore the chemical composition, sources and physical and chemical formation processes of prominent components. The results showed that five trace elements (Mn, Cu, As, Zn and Pb) had high enrichment in PM₁₀ and were closely related with anthropogenic combustion and vehicle emissions; organic and element carbon had a high correlation due to the same primary sources and similar evolution; nitrate dominated SNA (sulfate, nitrate, ammonium) and nitrate/sulfate ratios reached 2.35 on the polluted days owing to the significant contribution of motor vehicle emissions. Positive matrix factorization analysis indicated that secondary source, traffic, biomass burning, industry, coal combustion and crustal dust were the main sources of PM₁₀, contributing 32.5%, 20.9%, 15.0%, 13.9%, 9.4% and 8.3%, respectively; backward trajectories and potential source contribution function analysis showed that short-distance airflow was the dominant cluster and accounted for nearly 50% of total trajectories. The Weather Research and Forecasting model with Chemistry, with integrated process rate analysis, showed that dominant gas-phase reactions (heterogeneous reaction) during daytime (nighttime) in presence of ammonia led to a significant enhancement of nitrate in Zhuozhou, contributing 12.6 μg/m³ in episode 1 and 22.9 μg/m³ in episode 2. Full article

To better inform the public about ambient air quality and associated health risks and prevent cardiovascular and chronic respiratory diseases in Macau, the local government authorities apply the Air Quality Index (AQI) for air quality management within its jurisdiction. The application of AQI requires first determining the sub-indices for several pollutants, including respirable suspended particulates (PM₁₀), fine suspended particulates (PM_2.5), nitrogen dioxide (NO₂), ozone (O₃), sulfur dioxide (SO₂), and carbon monoxide (CO). Accurate prediction of AQI is crucial in providing early warnings to the public before pollution episodes occur. To improve AQI prediction accuracy, deep learning methods such as artificial neural networks (ANNs) and long short-term memory (LSTM) models were applied to forecast the six pollutants commonly found in the AQI. The data for this study was accessed from the Macau High-Density Residential Air Quality Monitoring Station (AQMS), which is located in an area with high traffic and high population density near a 24 h land border-crossing facility connecting Zhuhai and Macau. The novelty of this work lies in its potential to enhance operational AQI forecasting for Macau. The ANN and LSTM models were run five times, with average pollutant forecasts obtained for each model. Results demonstrated that both models accurately predicted pollutant concentrations of the upcoming 24 h, with PM₁₀ and CO showing the highest predictive accuracy, reflected in high Pearson Correlation Coefficient (PCC) between 0.84 and 0.87 and Kendall’s Tau Coefficient (KTC) between 0.66 and 0.70 values and low Mean Bias (MB) between 0.06 and 0.10, Mean Fractional Bias (MFB) between 0.09 and 0.11, Root Mean Square Error (RMSE) between 0.14 and 0.21, and Mean Absolute Error (MAE) between 0.11 and 0.17. Overall, the LSTM model consistently delivered the highest PCC (0.87) and KTC (0.70) values and the lowest MB (0.06), MFB (0.09), RMSE (0.14), and MAE (0.11) across all six pollutants, with the lowest SD (0.01), indicating greater precision and reliability. As a result, the study concludes that the LSTM model outperforms the ANN model in forecasting air pollutants in Macau, offering a more accurate and consistent prediction tool for local air quality management. Full article

Air pollution monitoring in Korea has not yet been implemented in agricultural areas. Documenting air quality in purely agricultural areas is inherently valuable. This study compares agricultural air quality with urban air quality during two periods: (1) the entire measurement period and (2) high-PM episodes. To ensure broad spatial coverage, eight monitoring stations were installed in Yeoju, Nonsan, Naju, Gimhae, Hongcheon, Danyang, Muan, and Sangju. Real-time measurements of PM₁₀, PM_2.5, SO₂, and NOx were conducted continuously from March 2023 to December 2024. Over the entire measurement period, PM concentrations were similar in both agricultural and urban areas, but gaseous pollutants were lower in agricultural areas. PM levels were higher in agricultural areas during summer, whereas urban areas showed higher concentrations in other seasons. During high-PM episodes (29 days), all pollutants were significantly higher in urban areas, with PM_2.5 showing a greater difference than PM₁₀. Diurnal variations revealed that PM₁₀, PM_2.5, and NO₂ peaked in the morning and reached their lowest levels around 3 PM, with urban levels consistently higher than those in agricultural areas. SO₂ showed a different pattern, reaching its lowest concentration at 6 AM and peaking at noon in urban areas and at 6 PM in agricultural areas. This pattern closely followed temperature and wind speed variations. Full article

Air pollution, particularly fine particulate matter (PM_2.5), poses significant environmental and public health challenges in South Korea. The National Institute of Environmental Research (NIER) currently relies on numerical models such as the Community Multiscale Air Quality (CMAQ) model for PM_2.5 forecasting. However, these models exhibit inherent uncertainties due to limitations in emission inventories, meteorological inputs, and model frameworks. To address these challenges, this study evaluates and compares the forecasting performance of two alternative models: Long Short-Term Memory (LSTM), a deep learning model, and Seasonal Auto Regressive Integrated Moving Average with Exogenous Variables (SARIMAX), a statistical model. The performance evaluation was focused on Seoul, South Korea, and took place over different forecast lead times (D00–D02). The results indicate that for short-term forecasts (D00), SARIMAX outperformed LSTM in all statistical metrics, particularly in detecting high PM_2.5 concentrations, with a 19.43% higher Probability of Detection (POD). However, SARIMAX exhibited a sharp performance decline in extended forecasts (D01–D02). In contrast, LSTM demonstrated relatively stable accuracy over longer lead times, effectively capturing complex PM_2.5 concentration patterns, particularly during high-concentration episodes. These findings highlight the strengths and limitations of statistical and deep learning models. While SARIMAX excels in short-term forecasting with limited training data, LSTM proves advantageous for long-term forecasting, benefiting from its ability to learn complex temporal patterns from historical data. The results suggest that an integrated air quality forecasting system combining numerical, statistical, and machine learning approaches could enhance PM_2.5 forecasting accuracy. Full article

Understanding the correlation between PM_2.5 and O₃ is critical for complex air pollution control. This study comprehensively analyzed PM_2.5 and O₃ pollution characteristics, uncovered spatiotemporal variations in their correlation, and investigated the driving mechanisms of their association in Dongying, a typical petrochemical city in China’s Bohai Bay region. Results showed that PM_2.5–O₃ correlation in Dongying exhibited significant seasonal variations, spatial patterns, and concentration threshold effects from 2017 to 2023. PM_2.5 and O₃ showed strong positive correlations in summer, negative in winter, and weak positive in spring/autumn, with strongest links in western areas. The strongest positive PM_2.5–O₃ correlation occurred in summer when PM_2.5 ≤ 35 μg·m⁻³ and O₃ >160 μg·m⁻³, while the strongest negative correlation was exhibited in winter with PM_2.5 > 75 μg·m⁻³ and O₃ ≤ 100 μg·m⁻³. Meteorological conditions (T > 20 °C, RH < 30%, wind speed < 1.73 m/s, O_x > 125 μg·m⁻³) and non-sea-breeze periods enhanced the PM_2.5–O₃ positive correlation. During the four typical pollution episodes, the positive PM_2.5–O₃ correlation in summer was propelled by synchronous increases in O₃ and secondary components via shared precursors. In autumn, strong positivity resulted from secondary component–O₃ correlations (r > 0.7) and dominance of secondary formation in PM_2.5. In winter, the negative correlation stemmed from primary emissions inhibiting photochemistry. Random forest analysis showed that O_x, RH, and T drove positive PM_2.5–O₃ correlation via photochemistry in summer, whereas winter primary emissions and NO titration caused negative correlation. This study offers guidance for the collaborative PM_2.5 and O₃ control in the petrochemical cities of the Bay region. Full article

Although the benefits of trees in cities are of great variety and value, attention must also be paid to the consequences for public health of the presence of pollen aeroallergens in the atmosphere, which are likely to interact with air pollutants, influencing the alteration of the immune system, facilitating allergic reactions, and enhancing the symptoms of asthmatic patients. This study analyses (using multiple linear regression calculations performed with the data analysis tool Statgraphics Centurion 19) the interaction of the concentration of six types of tree pollen (Cupressaceae, Olea, Platanus, Pinus, Ulmus, and Populus) and six atmospheric pollutants (O₃, PM₁₀ and PM_2.5, NO₂, CO, and SO₂), on asthma care episodes in the Community of Madrid (CAM). In most of the calculated equations, the adjusted R² value is higher than 30%, and in all cases, the P-value of the models obtained is lower than 0.0001. Therefore, almost all models obtained in the study period for asthma are statistically significant. Olea is the pollen type most frequently associated with asthma (followed by Pinus and Populus), in all the years studied. In the same period, O₃ is the most common air pollutant in the equations obtained for asthma. Stronger interrelations with asthma are generally found in more urban municipalities. Full article

Air pollution in Chiang Mai during the dry winter season is extremely severe. During this period, high levels of fine particles are primarily generated by open biomass burning in Thailand and neighboring countries. In this study, ambient VOC(Volatile Organic Compounds) samples were collected using an adsorbent tube from 13 March to 26 March 2024, with careful consideration of sampling uncertainties to ensure data reliability. Furthermore, while interannual variability exists, the findings reflect atmospheric conditions during this specific period, allowing for an in-depth VOC assessment. A comprehensive approach to VOCs was undertaken, including benzene, toluene, ethylbenzene, m,p-xylene (BTEX); biogenic volatile organic compounds (BVOCs); and carbonyl compounds. Regression analysis was performed to analyze the correlation between isoprene concentrations and wind direction. The results showed a significant variation in isoprene levels, indicating their high concentrations due to biomass burning originating from northern areas of Chiang Mai. The emission sources of BTEX and carbonyl compounds were inferred through their ratio analysis. Additionally, correlation analyses between PM_2.5, BTEX, and carbonyl compounds were conducted to identify common emission pathways. The ratio of BTEX among compounds suggested that long-range pollutant transport contributed more significantly than local traffic emissions. Carbonyl compounds were higher during the episode period, which was likely due to local photochemical reactions and biological contributions. Previous studies in Chiang Mai have primarily focused on PM_2.5, whereas this study examined individual VOC species, their temporal trends, and their interrelationships to identify emission sources. Full article

This study aims to assess the short-term effects of extreme heat, cold, and air pollution episodes on excess mortality from natural causes in Luxembourg over 1998–2023. Using a high-resolution dataset from downscaled and bias-corrected temperature (ERA5) and air pollutant concentrations (EMEP MSC-W), weekly mortality p-scores were linked to environmental episodes. A distributional regression approach using a logistic distribution was applied to model the influence of environmental risks, capturing both central trends and extreme values of excess mortality. Results indicate that extreme heat, cold, and fine particulate matter (PM_2.5) episodes significantly drive excess mortality. The estimated attributable age-standardized mortality rates are 2.8 deaths per 100,000/year for extreme heat (95% CI: [1.8, 3.8]), 1.1 for extreme cold (95% CI: [0.4, 1.8]), and 6.3 for PM_2.5 episodes (95% CI: [2.3, 10.3]). PM_2.5-related deaths have declined over time due to the reduced frequency of pollution episodes. The odds of extreme excess mortality increase by 1.93 times (95% CI: [1.52, 2.66]) per extreme heat day, 3.49 times (95% CI: [1.77, 7.56]) per extreme cold day, and 1.11 times (95% CI: [1.04, 1.19]) per PM_2.5 episode day. Indicators such as return levels and periods contextualize extreme mortality events, such as the p-scores observed during the 2003 heatwave and COVID-19 pandemic. These findings can guide public health emergency preparedness and underscore the potential of distributional modeling in assessing mortality risks associated with environmental exposures. Full article