Township-level PM2.5 Estimation in China

PM pollution in Chinese cities is among the worst in the world after decades of rapid economic growth with loose environmental regulation (Zhang et al., 2012). The Chinese ambient air quality standards used to regulate inhalable particles (PM10) only. Not until very recently, China’s PM2.5 pollution has drawn national and international attention, especially after a series of “airpocalypse” events. These events have triggered the Chinese government to tackle the serious fine particle pollution problem by using information strategy. The information disclosure about PM2.5 pollution also helps Chinese citizens to mitigate their exposure to fine particles and reduce health risks by adopting avoidance behavior like avoiding outdoor activity, wearing masks.


The objective of this paper is to assess China’s potential population exposure to PM2.5, map its spatiotemporal variability, and simulate the implication of various PM2.5 concentration standards. To this end, we relate Moderate Resolution Imaging Spectroradiometer (MODIS) Aerosol Optical Depth (AOD) retrievals to ground-based PM2.5 observations. We employ block cokriging to improve the spatial interpolation and extrapolation of PM2.5 distribution (Goovaerts, 1997). In addition, we use the sub-district level population data to estimate and map potential population exposure to PM2.5 pollution in China at a fine scale.


We find that Chinese people are potentially exposed to severe PM2.5 pollution. The population-weighted annual average PM2.5 concentration and number of polluted days reached 68.3 μg/m3 and 113 days, respectively, during April 08, 2013 and April 07, 2014. According to China's ambient air quality standards with daily average PM2.5 concentrations of 75 μg/m3, about 223 million people lived in a polluted environment for longer than half a year. The exposed population increases to 1,322 million, or 98.7% of China's total population, if the World Health Organization’s (WHO) recommended standard (25 μg/m3) is adopted. We also find significant spatiotemporal variability in the pollution pattern. Spatially, pollution is generally worse for the cities in the north, in the inland areas, or in the plain and basin regions. Temporally, PM2.5 concentrations reach the peak during fall and winter for both North and South China.



To the best of our knowledge, we are the first to assess China’s potential population exposure to PM2.5 on a daily basis at the sub-district level. Although the similar question has been explored in previous studies, they mostly focus on one city like Beijing or Tianjin (Zhang et al., 2013; Chen et al., 2010). In comparison, our assessment covers the whole country at a very fine scale, which is made possible by the recently available data and methods.


Reported by Pengpai in Chinese

Our paper for the project has been published in IJERRH in 2018.
Long et al 2018 IJERPH_PM2.5.pdf
Adobe Acrobat Document 3.5 MB
The map for downloading
JPG Image 3.9 MB
The inventory of estimated exposure for all Chinese cities
In English
Microsoft Excel Table 20.8 KB
The slides for our study
Adobe Acrobat Document 2.7 MB

Adobe Acrobat Document 297.7 KB
Adobe Acrobat Document 834.7 KB

More details in the working paper 45 of BCL.

Media Coverage by Tencent