Beyond Residential Ambient Concentrations: Quantifying Exposure Error and Advancing Personal PM 2.5 Prediction with a Scalable Modeling Framework
Environmental Science & Technology · 10.02.2026 · 60 (5) · 4271-4278
DOI: 10.1021/acs.est.5c14322 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Science & Technology, online erschienen am 26.01.2026. Verfasst von Xinjie Dai und 10 weiteren Autor:innen. Beteiligte Einrichtung: School of Public Health (Shenzhen), Shenzhen Campus of Sun Yat-Sen University, Shenzhen, Guangdong 518107, China. Hauptthemen laut MEDLINE: Particulate Matter, Environmental Exposure. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: Fundamental Research Funds for the Central Universities, National Natural Science Foundation of China, Shenzhen Key Laboratory of Pathogenic Microbes and Biosafety.
Abstract (Originalsprache)
Accurate assessment of personal PM2.5 exposure is essential but challenging in large-scale epidemiology, as conventional residential ambient data often lead to exposure misclassification. This study aimed to quantify errors in ambient data proxies and develop a scalable modeling framework for personal exposure prediction using readily available data. We conducted a panel study with 12 adults from three diverse Chinese cities, collecting 4571 person-hours of personal PM2.5 measurements. These were compared against three ambient data sources to quantify relative errors. We developed a modeling framework integrating ambient concentrations, meteorological variables, and basic personal characteristics, incorporating systematic preprocessing, feature engineering, variable selection, and multialgorithm comparison optimized through hyperparameter tuning and cross-validation. Results showed substantial personal-ambient exposure discrepancies, with relative errors of 39-48% for the daily average level. The framework successfully predicted personal exposure, with a Random Forest model using daily monitoring-station data achieving the highest performance (R2 = 0.87). SHAP analysis identified ambient PM2.5 as the dominant predictor, with personal traits and meteorology also contributing significantly. This work provides a validated, end-to-end modeling framework that moves beyond ambient proxies, offering a standardized workflow to refine personal exposure estimates in large cohorts and enhance the validity of air pollution health studies.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- School of Public Health (Shenzhen), Shenzhen Campus of Sun Yat-Sen University, Shenzhen, Guangdong 518107, China.
- Global Health Institute and the Nicholas School of Environment, Duke University, Durham, North Carolina 27708, United States.
- Intelligent Sensing and Proactive Health Research Center, Sun Yat-Sen University, Shenzhen, Guangdong 518107, China.
- Shenzhen Key Laboratory of Pathogenic Microbes and Biosafety, Shenzhen, Guangdong 518107, China.
ORCID-Profile: Junfeng Jim Zhang, Jianbang Xiang
Förderung
- Fundamental Research Funds for the Central Universities
- National Natural Science Foundation of China
- Shenzhen Key Laboratory of Pathogenic Microbes and Biosafety
Zitieren
APA
Dai, X., Zhang, R., Li, Q., Wang, C., Hu, L., Fang, X., Yang, C., Zhou, Y., Zhou, R., Zhang, J. J., & Xiang, J. (2026). Beyond Residential Ambient Concentrations: Quantifying Exposure Error and Advancing Personal PM 2.5 Prediction with a Scalable Modeling Framework. Environmental Science & Technology, 60(5), 4271-4278. https://doi.org/10.1021/acs.est.5c14322
Vancouver
Dai X, Zhang R, Li Q, Wang C, Hu L, Fang X, et al. Beyond Residential Ambient Concentrations: Quantifying Exposure Error and Advancing Personal PM 2.5 Prediction with a Scalable Modeling Framework. Environmental Science & Technology. 2026;60(5):4271-4278. doi:10.1021/acs.est.5c14322
BibTeX
@article{Dai2026,
author = {Dai, Xinjie and Zhang, Ruitong and Li, Qing and Wang, Chunliang and Hu, Linmin and Fang, Xixian and Yang, Chunhui and Zhou, Yuyan and Zhou, Runhua and Zhang, Junfeng Jim and Xiang, Jianbang},
title = {Beyond Residential Ambient Concentrations: Quantifying Exposure Error and Advancing Personal PM 2.5 Prediction with a Scalable Modeling Framework},
journal = {Environmental Science & Technology},
year = {2026},
volume = {60},
number = {5},
pages = {4271--4278},
doi = {10.1021/acs.est.5c14322}
}
EUVIMED