China’s National Environmental Monitoring Center (CNEMC) now has real-time, hourly air quality data for 113 major Chinese cities, available here. Concentrations in (mg/m3) are reported for major pollutants sulfur dioxide (SO2), nitrogen dioxide  (NO2), and particulate matter (PM10, particles with a diameter of 10 microns or less).  This is a major improvement from previous data, which only reported air quality in terms of the Air Pollution Index (API).  A major complaint of the API is that its air quality categories (i.e. “slightly polluted,” “heavily polluted,” etc.) are only based on the criteria pollutant with the greatest concentration (more often than not PM10 in the smoggy skies of Beijing, for instance) averaged over a 24-hour period.  See my friend Vance Wagner’s lucid explanation of how the API is calculated in China. He has also written a few posts on the new system as well as translations of the explanation pages available in the website’s introduction.

Therefore, this new web application on the CNEMC website (available in Chinese online), represents a significant improvement by the Chinese government to be more transparent about air quality data.  Pollutant data are available for all 31 Chinese provinces and municipalities (Beijing, Shanghai, Tianjin, and Chongqing), with some provinces containing multiple datapoints from a few cities and air monitoring stations.  The data points, however, are not comprehensive: the stations listed for Beijing only include 8 of the 27 monitoring stations used to calculate Beijing’s API.  There’s also a question about of how the data compare with other data sources in China, such as the US Embassy‘s measurements.

I’d like to use this data for my research, but still haven’t managed to find an archive or download option. If you come across either or have any other ideas about how I might collect the data for further analysis, please drop me a line.

— Update — Jan. 7, 2011

Thank you to all who provided comments and questions that got me to take a closer look at the data available on the site.  Here are a few more observations:

– Because the application uses Microsoft Silverlight, it doesn’t look like there is capacity at the moment to download or scrape data from the site.  It’s a little frustrating, considering the data gold mine it presents.

– Class I, II, and III air quality standards for the criteria air pollutants SO2, NO2, and PM10 are in the application’s introduction (空气质量监测系统介绍), which helps to provide some explanation of what the pollutant concentrations might mean.  Here’s a table I made that compares the Chinese air quality standards to their normalized scores on the Air Pollution Index (API):

Pollutant Time period Class 1


Class 2


Class 3

(Slightly-lightly polluted)

mg/m3 API mg/m3 API mg/m3 API
SO2 Yearly 0.02 <50 0.06 50-100 0.10 50-100
Daily 0.05 50 0.15 100 0.25 100-200
Hourly 0.15 100 0.50 100-200 0.70 100-200
PM10 Yearly 0.04 <50 0.10 50-100 0.15 100
Daily 0.05 50 0.15 100 0.25 100-200
NO2 Yearly 0.04 <50 0.08 50 0.08 50
Daily 0.08 50 0.12 100 0.12 100
Hourly 0.12 100 0.24 100-200 0.24 100-200

Table 1. Hourly, Daily, and Yearly Air Pollutant Concentrations and Corresponding Air Pollution Index (API) Scores


On China’s API, anything below 100 is considered a ‘Blue-Sky Day.’  Take notice that by default the graphs for the hourly pollutant data has a red line for the Class 2 standards for each pollutant:


So it looks like at some times of the day, hourly exposure to some of these pollutants could fall into the ‘lightly polluted’ or ‘moderately lightly polluted categories’ (API 100-200).  The API is, of course, calculated using the 24-hour averages of the pollutants.

– At any rate, this site seems to be a step forward toward greater transparency on environmental data in China. The Institute of Public and Environmental Affairs (IPE) and the Natural Resources Defense Council just released the second edition of the Pollution Information Transparency Index (PITI; Chinese only).  They saw modest gains in transparency of environmental data in 113 Chinese cities, although overall China still has a ways to go.

– More to come later, when I figure out a way to download some of the data to compare to satellite information.