Related Experiment Video
Updated: Jan 9, 2026
08:17
Generation of a Chronic Obstructive Pulmonary Disease Model in Mice by Repeated Ozone Exposure
Published on: August 25, 2017
11.4K
Source-resolved PM2.5 oxidative potential predicts global mortality burden: A machine learning approach.
1Department of Environmental Engineering, Pusan National University, 2, Busandaehak-ro 63beon-gil, Geumjeong-gu, Busan, 46241, Republic of Korea.
Journal of Environmental Management
|December 4, 2025
Summary
Source-resolved oxidative potential (OPv) from fine particulate matter (PM2.5) is a better predictor of mortality than PM2.5 mass alone. This study developed global OPv source-resolved functions to improve air quality risk assessment and policy.
Area of Science:
- Environmental Health Sciences
- Atmospheric Chemistry
- Toxicology
Background:
- Fine particulate matter (PM2.5) oxidative potential (OPv) is a key health indicator.
- Existing concentration-response functions (CRFs) for PM2.5 mass lack source-specific data, limiting global health risk assessments.
- Source-resolved OPv (OPv^source) data is crucial for accurate epidemiological studies and targeted interventions.
Purpose of the Study:
- To develop global source-resolved OPv CRFs (OPv^source CRFs) for major emission sectors.
- To quantify the contribution of different emission sources to global OPv.
- To evaluate the predictive power of OPv^source for PM2.5-attributable mortality compared to PM2.5 concentration and OPv^PM2.5.
Main Methods:
- Integrated machine learning, statistical techniques, and extensive datasets on PM2.5 concentrations, source apportionment (50 countries), and measured OPv (>10,000 measurements in 29 countries).
- Developed intrinsic OP per unit mass (OPm) for energy (ENE), transportation (TRO), industry (IND), and agriculture/residential combustion (AGS + RCO) sectors.
- Applied developed CRFs to 2017 PM2.5 source data across 203 countries and used Poisson regression to assess mortality prediction.
Main Results:
- Distinct OPm values were identified for major PM2.5 emission sectors, with ENE > TRO > IND > AGS + RCO.
- The global average OPv^source was estimated at 0.78 nmol min⁻¹ m⁻³ (95% CI: [0.39, 1.2]), with ENE (33%) and AGS + RCO (31%) as primary contributors.
- OPv^source significantly outperformed PM2.5 concentration and OPv^PM2.5 in predicting PM2.5-attributable mortality.
Conclusions:
- Source-resolved OPv is a more accurate and critical metric for assessing PM2.5 health risks than previously used metrics.
- The developed global OPv^source CRFs provide a robust tool for national and international air quality management.
- Findings support the implementation of targeted policies to reduce the health burden associated with specific PM2.5 emission sources.
