A hybrid model for estimating the number concentration of ultrafine particles based on machine learning algorithms in central Taiwan

• Published in: Environment International Vol. 175; p. 107937
• Main Authors: Jung, Chau-Ren, Chen, Wei-Ting, Young, Li-Hao, Hsiao, Ta-Chih
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01.05.2023; Elsevier BV; Elsevier
• Subjects: Air Pollutants; Air Pollutants - analysis; Air Pollution; Air Pollution - analysis; cost effectiveness; environment; Environmental Monitoring; Environmental sciences; Estimation model; Feature importance; GE1-350; land use; Machine Learning; Meteorological variables; Particle Size; Particulate Matter; Particulate Matter - analysis; Satellite-based measurement; satellites; Taiwan; temperature; traffic; Ultrafine particles; wind; Taiwan; Feature importance; Ultrafine particles; Meteorological variables; Estimation model; Satellite-based measurement; Machine learning
• ISSN: 0160-4120; 1873-6750; 1873-6750
• DOI: 10.1016/j.envint.2023.107937

[Display omitted] •In-situ long-term measurements of ultrafine particles (UFP) were conducted in central Taiwan.•XGBoost model outperformed random forest and deep neural network.•The training and cross-validation R2 (nRMSE) were 0.99 (6.5%) and 0.78 (31.0%), respectively.•Surface pressure and traffic-related variables were important predictors.•MAIAC AOD and AE were not strong predictors for UFP. Modeling is a cost-effective measure to estimate ultrafine particle (UFP) levels. Previous UFP estimates generally relied on land-use regression with insufficient temporal resolution. We carried out in-situ measurements for UFP in central Taiwan and developed a model incorporating satellite-based measurements, meteorological variables, and land-use data to estimate daily UFP levels at a 1-km resolution. Two sampling campaigns were conducted for measuring hourly UFP concentrations at six sites between 2008–2010 and 2017–2021, respectively, using scanning mobility particle sizers. Three machine learning algorithms, namely random forest, eXtreme gradient boosting (XGBoost), and deep neural network, were used to develop UFP estimation models. The performances were evaluated with a 10-fold cross-validation, temporal, and spatial validation. A total of 1,022 effective sampling days were conducted. The XGBoost model had the best performance with a training coefficient of determination (R2) of 0.99 [normalized root mean square error (nRMSE): 6.52%] and a cross-validation R2 of 0.78 (nRMSE: 31.0%). The ten most important variables were surface pressure, distance to the nearest road, temperature, calendar year, day of the year, NO2, meridional wind, the total length of roads, PM2.5, and zonal wind. The UFP levels were elevated along the main roads across different seasons, suggesting that traffic emission is an important contributor to UFP. This hybrid model outperformed prior land use regression models and thus can provide more accurate estimates of UFP for epidemiological studies.