Application of machine learning in atmospheric pollution research: A state-of-art review

The Science of The Total Environment, Год журнала: 2023, Номер 910, С. 168588 - 168588

Опубликована: Ноя. 18, 2023

Язык: Английский

---

Clarifying Relationship between PM2.5 Concentrations and Spatiotemporal Predictors Using Multi-Way Partial Dependence Plots

Remote Sensing, Год журнала: 2023, Номер 15(2), С. 358 - 358

Опубликована: Янв. 6, 2023

Atmospheric fine particles (PM2.5) have been found to be harmful the environment and human health. Recently, remote sensing technology machine learning models used monitor PM2.5 concentrations. Partial dependence plots (PDP) were explore meteorology mechanisms between predictor variables concentration in “black box” models. However, there are two key shortcomings original PDP. (1) it calculates marginal effect of feature(s) on predicted outcome a model, therefore some local effects might hidden. (2) requires that for which partial is computed not correlated with other features, otherwise estimated feature has great bias. In this study, PDP’s analyzed. Results show contradictory correlation temperature can given by Furthermore, spatiotemporal heterogeneity PM2.5-AOD relationship cannot displayed well The drawbacks PDP make unsuitable exploring large-area effects. To resolve above issue, multi-way recommended, characterize how concentrations changed temporal spatial variations major meteorological factors China.

Язык: Английский

---

Key drivers of the oxidative potential of PM2.5 in Beijing in the context of air quality improvement from 2018 to 2022

Environment International, Год журнала: 2024, Номер 187, С. 108724 - 108724

Опубликована: Май 1, 2024

The mass concentration of atmospheric particulate matter (PM) has been continuously decreasing in the Beijing-Tianjin-Hebei region. However, health endpoints do not exhibit a linear correlation with PM concentrations. Thus, it is urgent to clarify prior toxicological components further improve air quality. In this study, we analyzed long-term oxidative potential (OP) water-soluble PM2.5, which generally considered more effective assessing hazardous exposure Beijing from 2018 2022 based on dithiothreitol assay and identified crucial drivers OP PM2.5 online monitoring pollutants, receptor model, random forest (RF) model. Our results indicate that dust, traffic, biomass combustion are main sources Beijing. complex interactions dust particles, black carbon, gaseous pollutants (nitrogen dioxide sulfur dioxide) factors driving evolution, particular, leading abnormal rise 2022. data shows higher observed winter spring compared summer autumn. diurnal variation characterized by declining trend 0:00 14:00 an increasing 23:00. spatial was as lower than Shijiazhuang, while Zhenjiang Haikou, primarily influenced distribution carbon. significance identifying key influencing provide new insights for advancing quality improvement efforts focus safeguarding human

Язык: Английский

---

Elucidating ozone and PM_2.5 pollution in the Fenwei Plain reveals the co-benefits of controlling precursor gas emissions in winter haze

Atmospheric chemistry and physics, Год журнала: 2023, Номер 23(6), С. 3595 - 3607

Опубликована: Март 23, 2023

Abstract. The Fenwei Plain, home to 50 million people in central China, is one of the most polluted regions China. In 2018, Plain was designated as three key for “Blue Sky Protection Campaign”, along with Beijing–Tianjin–Hebei (BTH) and Yangtze River Delta (YRD) regions. However, compared BTH YRD, our understanding current status air pollution limited partly due a lack detailed analysis transformation from precursor gases secondary products including organic aerosol (SOA) ozone. Through 7 years (2015–2021) surface monitoring pollutants Xi'an, largest city we show that roughly two-thirds days exceeded either PM2.5 or O3 level-1 quality standard, highlighting severity pollution. Moreover, an increase winter haze also revealed, constantly elevated reactive oxygenated volatile compounds (OVOCs), particular formaldehyde, ozone formation potential over µg m−3, combination reduced NO2. abrupt decrease NO2, observed during lockdown 2020, provided real-world evidence control measures, targeting only NOx (70 % on average), were insufficient reduce because OVOCs remained high compound (VOC)-limited regime. Model simulation results showed NO2 reduction 20 %–70 %, self-reaction rate between peroxy radicals, pathway SOA formation, intensified by up 75 while further VOCs > %. Therefore, synergic can be achieved through more aggressive their gases. This study elucidates revealing general trend increasing pollution, i.e., haze. Controlling gas emissions anticipated curb both which will benefit not just but other

Язык: Английский

---

An intercomparison of weather normalization of PM2.5 concentration using traditional statistical methods, machine learning, and chemistry transport models

npj Climate and Atmospheric Science, Год журнала: 2023, Номер 6(1)

Опубликована: Дек. 20, 2023

Abstract Traditional statistical methods (TSM) and machine learning (ML) have been widely used to separate the effects of emissions meteorology on air pollutant concentrations, while their performance compared chemistry transport model has less fully investigated. Using Community Multiscale Air Quality Model (CMAQ) as a reference, series experiments was conducted comprehensively investigate TSM (e.g., multiple linear regression Kolmogorov–Zurbenko filter) ML random forest extreme gradient boosting) approaches in quantifying trends fine particulate matter (PM 2.5 ) during 2013−2017. evaluation metrics suggested that can explain variations PM with highest from ML. The showed insignificant differences ( p > 0.05) for both emission-related $${{\rm{PM}}}_{2.5}^{{\rm{EMI}}}$$ PM 2.5 EMI meteorology-related components between TSM, ML, CMAQ modeling results. estimated least difference CMAQ. Considering medium computing resources low biases, method is recommended weather normalization . Sensitivity analysis further optimized hyperparameters exclusion temporal variables produce reasonable results

Язык: Английский

---

Increased contribution to PM2.5 from traffic-influenced road dust in Shanghai over recent years and predictable future

Environmental Pollution, Год журнала: 2022, Номер 313, С. 120119 - 120119

Опубликована: Сен. 16, 2022

Язык: Английский

---

Multiple driving factors and hierarchical management of PM2.5: Evidence from Chinese central urban agglomerations using machine learning model and GTWR

Urban Climate, Год журнала: 2022, Номер 46, С. 101327 - 101327

Опубликована: Ноя. 7, 2022

Язык: Английский

---

A Visibility-Based Historical PM2.5 Estimation for Four Decades (1981–2022) Using Machine Learning in Thailand: Trends, Meteorological Normalization, and Influencing Factors Using SHAP Analysis

Aerosol and Air Quality Research, Год журнала: 2025, Номер 25(1-4)

Опубликована: Март 27, 2025

Abstract Introduction PM 2.5 pollution is a significant environmental and health concern in Thailand, with levels intensifying during the dry season. However, lack of long-term PM2.5 data limits understanding historical trends meteorological influences. Objective This study aims to reconstruct from 1981 2022 analyze influence various contributing factors across six key provinces Thailand: Chiang Mai (CM), Lampang (LP), Khon Kaen (KK), Bangkok (BK), Chonburi (CB), Songkhla (SK). Methods A Light Gradient Boosting Machine (LightGBM) model was developed using aerosol-related variables Thai Meteorological Department MERRA-2. The trained on spanning 2012–2022, depending availability for each province. Model performance evaluated diurnal, monthly, annual scales then used reconstruction data. SHAP analysis determine important predictor affecting prediction. Results LightGBM accurately predicted all provinces, showing better daily prediction than hourly accuracy higher clean hours haze hours. Good agreement between observed found different time (diurnal, annually). CM shows non-significant trend, limiting insights into effects, while LP exhibits decreases PM2.5_emis, indicating positive weather impacts air quality. In contrast, regions like KK, BK, CB display worsening influences, or increasing despite declines _emis. SK, removing effects reveals decreasing underscoring critical role meteorology. identified visibility, gridded , specific humidity at 2 m as common over along additional that were not consistent provinces. Conclusion effectively reconstructs provides insight influences Based findings study, some policy implications have also been provided. Graphical abstract

Язык: Английский

---

Black carbon in urban Jinan: variations, health risks, and driving factors analyzed with machine learning

Journal of Environmental Sciences, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Язык: Английский

---

Machine learning-based quantification and separation of emissions and meteorological effects on PM2.5 in Greater Bangkok

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 28, 2025

Язык: Английский

---