Here we share some demonstrative and tutorial codes which were used to accomplish our research on ozone-exposure-associated mortality.
The research questions are: i) how ambient ozone change with space and time over the past 30 years, 1990-2019, ii) how urban and rural populations are exposed to different levels of ozone pollution, and iii) how manu premature deaths are caused from long-term ozone exposure.
The codes are classified into 4 categories, as listed below:
Analysis:
A1-1 Mortality estimation by curved risk model
A2-1 Change rate decomposition: Mortality 1990
A2-2 Change rate decomposition: Population growth
A2-3 Change rate decomposition: Population ageing
A2-4 Change rate decomposition: Baseline mortality change
A2-5 Change rate decomposition: Exposure increasing
A2-6 Change rate decomposition: Urbanisation migration
A2-7 Change rate decomposition: Summing up
Tutorial:
T1-1 Meta-analysis
T2-1 Mortality estimation by curved risk model
T2-2 Mortality estimation by linear risk model
Figures:
F1 Ambient ozone mapping
F2 Longitudinal ozone trends
F3 Excess mortality mapping
F4 Mortality piling-up
F5 Mortality sorting
F6 Change rate decomposition
Extended Analyses: EX1 Literature external validation EX2 Eagle-eye of longitudinal variability EX3 Population density projection
NATSUSTAIN-22113342B (R1) revision submitted on 12 Feb 2023 Codes are updated into v2.1.0 accordingly
The original ambient ozone database can be found at:
- The Bayesian neural network downscaled (BNNDv1) dataset: Sun et al. Spatial Resolved Surface Ozone with Urban and Rural Differentiation during 1990–2019: A Space–Time Bayesian Neural Network Downscaler. Environ. Sci. Technol. 2022, 56, 11, 7337–7349. https://doi.org/10.1021/acs.est.1c04797.
- Cluster-enhanced ensemble learning (CEML) trained dataset: Liu et al. Cluster-Enhanced Ensemble Learning for Mapping Global Monthly Surface Ozone From 2003 to 2019. https://doi.org/10.1029/2022GL097947.
- Ensemble-learning-based multi-data fusion for Tracking Air Pollution in China (TAP): Xue et al. Estimating Spatiotemporal Variation in Ambient Ozone Exposure during 2013–2017 Using a Data-Fusion Model. Environ. Sci. Technol. 2020, 54, 23, 14877–14888. https://doi.org/10.1021/acs.est.0c03098.