Premature Mortality Attributable to Particulate Matter in China: Source Contributions and Responses to Reductions.
Jianlin HuLin HuangMindong ChenHong LiaoHongliang ZhangShu-Xiao WangQiang ZhangQi YingPublished in: Environmental science & technology (2017)
Excess mortality (ΔMort) in China due to exposure to ambient fine particulate matter with aerodynamic diameter ≤2.5 μm (PM2.5) was determined using an ensemble prediction of annual average PM2.5 in 2013 by the community multiscale air quality (CMAQ) model with four emission inventories and observation data fusing. Estimated ΔMort values due to adult ischemic heart disease, cerebrovascular disease, chronic obstructive pulmonary disease, and lung cancer are 0.30, 0.73, 0.14, and 0.13 million in 2013, respectively, leading to a total ΔMort of 1.3 million. Source-oriented CMAQ modeling determined that industrial and residential sources were the two leading sources of ΔMort, contributing to 0.40 (30.5%) and 0.28 (21.7%) million deaths, respectively. Additionally, secondary ammonium ion from agriculture, secondary organic aerosol, and aerosols from power generation were responsible for 0.16, 0.14, and 0.13 million deaths, respectively. A 30% ΔMort reduction in China requires an average of 50% reduction of PM2.5 throughout the country and a reduction by 62%, 50%, and 38% for the Beijing-Tianjin-Hebei, Jiangsu-Zhejiang-Shanghai, and Pearl River Delta regions, respectively. Reducing PM2.5 to the CAAQS grade II standard of 35 μg m-3 would only lead to a small reduction in mortality, and a more stringent standard of <15 μg m-3 would be needed for more remarkable reduction of ΔMort.
Keyphrases
- particulate matter
- air pollution
- lung function
- chronic obstructive pulmonary disease
- cardiovascular events
- drinking water
- risk factors
- climate change
- mental health
- water soluble
- coronary artery disease
- wastewater treatment
- heavy metals
- electronic health record
- deep learning
- young adults
- optical coherence tomography
- artificial intelligence
- convolutional neural network