Machine learning combined with the PMF model reveal the synergistic effects of sources and meteorological factors on PM2.5 pollution

Zhang, Zhongcheng; Xu, Bo; Xu, Weiman; Wang, Rui; Gao, Jie; Li, Yue; Li, Mei; Feng, Yinchang; Shi, Guoliang

doi:10.1016/j.envres.2022.113322

Cited by 42 publications

(7 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since 2022, the environmental scenarios of applying ML algorithms have been further expanded. For instance, ML algorithms have been widely used for improving the efficiency of environmental monitoring and policy-making [ 27 ], accounting carbon budget [ 33 , 34 ], decoupling the meteorological impact on air pollution [ 9 , 35 ], screening the new pollutants from a tremendous number of chemicals [ 36 ], predicting the health benefits through reducing pollution [ [37] , [38] , [39] , [40] , [41] , [42] ], identifying the impactors affecting the food chain or ecosystem [ 43 , 44 ], etc. Example ML algorithms used in environmental research include recurrent neural network (RNN) [ 45 ], convolutional neural network (CNN) [ 46 ], decision tree [ 47 ], support vector machine (SVM) [ 48 , 49 ], random forest (RF) [ 8 , 10 ], and artificial/deep neural network [ 22 ].…”

Section: Introductionmentioning

confidence: 99%

A new ChatGPT-empowered, easy-to-use machine learning paradigm for environmental science

An,

Li,

Huang

et al. 2024

Eco-Environment & Health

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A new ChatGPT-empowered, easy-to-use machine learning paradigm for environmental science

An,

Li,

Huang

et al. 2024

Eco-Environment & Health

View full text Add to dashboard Cite

“…5. Meteorological conditions [16][17][18][19][20] also play a crucial role in PM2.5 concentrations. Temperature influences the rate of the chemical reactions involved in the formation of secondary pollutants, including PM2.5.…”

Section: Introductionmentioning

confidence: 99%

Long-Term Forecasting of Air Pollution Particulate Matter (PM2.5) and Analysis of Influencing Factors

Zhang,

Sun,

Liu

et al. 2023

Sustainability

View full text Add to dashboard Cite

Long-term forecasting and analysis of PM2.5, a significant air pollution source, is vital for environmental governance and sustainable development. We evaluated 10 machine learning and deep learning models using PM2.5 concentration data along with environmental variables. Employing explainable AI (XAI) technology facilitated explainability and formed the basis for factor analysis. At a 30-day forecasting horizon, ensemble learning surpassed deep learning in performance, with CatBoost emerging as the top-performing model. For forecasting horizons of 90 and 180 days, Bi-SLTM and Bi-GRU, respectively, exhibited the highest performance. Through an analysis of influencing factors by SHAP, it was observed that PM10 exerted the greatest impact on PM2.5 forecasting. However, this effect was particularly pronounced at higher concentrations of CO. Conversely, at lower CO concentrations, the impact of increased PM10 concentrations on PM2.5 was limited. Hence, it can be inferred that CO plays a pivotal role in driving these effects. Following CO, factors such as “dew point” and “temperature” were identified as influential. These factors exhibited varying levels of linear correlation with PM2.5, with temperature showing a negative correlation, while PM10, CO, and dew point generally demonstrated positive correlations with PM2.5.

show abstract

“…can yield information on health risks and excess mortality assessments, as well as helping in source identification and apportionment. The source identification and apportionment of PM are usually performed using receptor models, including chemical mass balance (CMB) [42,43], principal component analysis (PCA) [44,45], and positive matrix factorization (PMF) [46,47]. According to the characteristics of different PM components, the contributions of different pollution sources have been obtained to offer a clear approach to PM control.…”

Section: Introductionmentioning

confidence: 99%

Characterization and Source Apportionment of PM in Handan—A Case Study during the COVID-19

Shu

Wang

et al. 2023

Atmosphere

View full text Add to dashboard Cite

Handan is a typical city affected by regional particulate pollution. In order to investigate particulate matter (PM) characterization, source contributions and health risks for the general populations, we collected PM samples at two sites affected by a pollution event (12–18 May 2020) during the COVID-19 pandemic and analyzed the major components (SNA, OCEC, WSIIs, and metal elements). A PCA-MLR model was used for source apportionment. The carcinogenic and non-carcinogenic risks caused by metal elements in the PM were assessed. The results show that the renewal of old neighborhoods significantly influences local PM, and primarily the PM10; the average contribution to PM10 was 27 μg/m3. The source apportionment has indicated that all other elements came from dust, except Cd, Pb and Zn, and the contribution of the dust source to PM was 60.4%. As PM2.5 grew to PM10, the PM changed from basic to acidic, resulting in a lower NH4+ concentration in PM10 than PM2.5. The carcinogenic risk of PM10 was more than 1 × 10−6 for both children and adults, and the excess mortality caused by the renewal of the community increased by 23%. Authorities should pay more attention to the impact of renewal on air quality. The backward trajectory and PSCF calculations show that both local sources and short-distance transport contribute to PM—local sources for PM10, and short-distance transport in southern Hebei, northern Henan and northern Anhui for PM2.5, SO2 and NO2.

show abstract

Machine learning combined with the PMF model reveal the synergistic effects of sources and meteorological factors on PM2.5 pollution

Cited by 42 publications

References 90 publications

A new ChatGPT-empowered, easy-to-use machine learning paradigm for environmental science

A new ChatGPT-empowered, easy-to-use machine learning paradigm for environmental science

Long-Term Forecasting of Air Pollution Particulate Matter (PM2.5) and Analysis of Influencing Factors

Characterization and Source Apportionment of PM in Handan—A Case Study during the COVID-19

Contact Info

Product

Resources

About