Predicting daily concentrations of nitrogen dioxide, particulate matter and ozone at fine spatial scale in Great Britain

Atmospheric Pollution Research

article

Weiyi Wang¹  Daniela Fecht¹  Sean Beevers¹  John Gulliver² 

[1] MRC Centre for Environment and Health, School of Public Health, Imperial College London;Centre for Environmental Health and Sustainability, George Davies Centre, University of Leicester, University Road

关键词: Air pollution;    Exposure assessment;    Spatio-temporal model;    Land use regression;    Geographic information system;    Epidemiology;   

DOI  :  10.1016/j.apr.2022.101506

学科分类：农业科学（综合）

来源: Dokuz Eylul Universitesi * Department of Environmental Engineering

【 摘 要 】

Short-term exposure studies have often relied on time-series of air pollution measurements from monitoring sites. However, this approach does not capture short-term changes in spatial contrasts in air pollution. To address this, models representing both the spatial and temporal variability in air pollution have emerged in recent years. Here, we modelled daily average concentrations of nitrogen dioxide (NO 2 ), particulate matter (PM 2.5 and PM 10 ) and ozone (O 3 ) on a 25 m grid for Great Britain from 2011 to 2015 using a generalised additive mixed model, with penalised spline smooth functions for covariates. The models included local-scale predictors derived using a Geographic Information System (GIS), daily estimates from a chemical transport model, and daily meteorological characteristics. The models performed well in explaining the variability in daily averaged measured concentrations at 48–85 sites: 63% for NO 2 , 77% for PM 2.5 , 80% for PM 10 and 85% for O 3 . Outputs of the study include daily air pollution maps that can be applied in epidemiological studies across Great Britain. Daily concentration values can also be predicted for specific locations, such as residential addresses or schools, and aggregated to other exposure time periods (including weeks, months, or pregnancy trimesters) to facilitate the needs of different health analyses.

【 授权许可】

CC BY