Follow-up: Just how does China’s air quality compare globally?

Thank you to all who read my previous post, “Chinese NGO releases Air Quality Transparency Index,” and to those have provided some excellent feedback and food for thought. In further exploring the link between information transparency and environmental performance, I made a statement that Chinese cities have a notorious reputation for being the world’s most polluted; however, data limitations could mean that there may be worse air quality out there.

This point has been made by me and several of my colleagues who are experts in China’s environment, but does the data exist to prove it?  Late last year, Canadian researchers van Donkelaar and Martin came out with a remarkable study that used satellite data to map global PM 2.5 (air particulates with a diameter of 2.5 microns or less; known to have serious health implications due to their ability to penetrate human lungs) concentrations.  I wrote a post about their work on this blog, and also included a screenshot of their map (Figure 1).  As one can clearly see, the darkest red areas corresponding to PM 2.5 concentrations in the range of 80 μg/m³ fall mostly in Eastern China, with some spots in Central Africa and India.  To give you a sense of how severe these concentrations are, the U.S. EPA sets a primary standard of 15 μg/m³ for an annual average.

Figure 1. Global estimates of PM 2.5 from 2001-2006. (Source: van Donkelaar and Martin, 2010).

But, as an astute China observer will know, a lot can happen in four years.  In particular, the Chinese government between 2006 and 2010 implemented its Eleventh Five-Year Plan, which included aggressive targets for energy intensity (20% reductions from 2005 levels by 2010) and sulfur dioxide (10% reductions from 2005 levels by 2010).  By the government’s accounts, China was able to meet these targets just in the nick of time.

Although I haven’t seen any more recent studies using satellite data to evaluate the impact of China’s 11th Five-Year Plan policy (which is also why I’ve chosen to work on this as part of my dissertation), one systematic evaluation of policy performance on a global level is the Environmental Performance Index (EPI), a policy tool developed by the Yale Center for Environmental Law and Policy and Columbia’s Center for International Earth Information Science Network (CIESIN). While I won’t go into detail about how the index is calculated, indicators such as urban particulates, ozone, and indoor air quality are used to evaluate country performance on air quality (although the ozone indicator was removed from 2008 to 2010).

I’ve pulled a few screenshots from the EPI website (Figures 2) to illustrate that in 2010, according to Yale and Columbia’s analysis, China did not perform the worst on air quality amongst the 163 countries analyzed.

Figure 2. 2010 EPI metrics for Air Pollution effects on Human Health. Source:

A few notes of caution: if you visit the EPI website, you’ll notice data and figures for both 2008 and 2010 are available, to allow for some comparison over time.  However, the indicators used to gauge Air Quality for Human Health change between 2008 and 2010, notably the “Local ozone” indicator is removed between years, which the authors of the report explain as an attempt to reduce the amount of modeled data and redundancy in the indicators. The 2010 EPI air quality indicators also include nitrogen-oxides and non-methane volatile-organic compounds.

Figure 3. How China compares with other countries in Asia and the Pacific in the 2010 EPI on Air Pollution. Source:

I’d suggest playing around with the EPI website and taking a look at neat visualizations such as Figure 3, which allows for different group comparisons based on region, national GDP, peer groups, and EPI rankings. Again, when compared to other countries in the Asia Pacific region, China did not perform the worst in 2010 on air pollution.  This, of course, does not mean that pollution was not the most severe in China in 2010 or some Chinese cities did not have the worst air quality in absolute terms of pollutant concentrations.  The EPI provides a snapshot in time of how close policies put in place to control air pollution were in terms of reaching targets. In 2010 the target for indoor air pollution was 0% of the population exposed; and an outdoor air pollutant concentration less than or equal to 20 μg/m³.

While this information may not wholly satisfy the answer to the question of whether China’s air pollution is the most severe, it does make the point that China’s air pollution control policies are having an impact on air quality performance.  And further, this performance ranks better than some other countries but still ranks quite low overall.

On another note, I’ve taken on the task of managing the 2012 version of the EPI.  After a pilot phase (2008) and a follow-up round (2010), we’re considering making changes to the EPI to make it more robust to provide the most accurate picture of performance possible.  We’re also considering drilling down into the air quality data and incorporating satellite measurements to provide a more detailed look at policy performance over a longer time frame.

If you know of any datasets we might be missing or have any ideas on how we can improve upon this work, please comment on this post or send me a line.

Facebook Twitter Email

, , , , ,

  • Graham Webster

    Do any of these satellite particulates data come in time series? Boy I would love to get my hands on that.

    UPDATE: I mean to say: data as granular as the image above, not so much the country data.

  • Angel Hsu

    Yes, the MODIS satellite (what van Donkelaar and Martin use) provides daily data from 2000. It’s tricky using satellite data though, you’ve got to know what you’re doing. My next post will outline some of these challenges, which I’m grappling with right now.

  • Graham Webster

    Yeah sounds like a learning curve. My dream dataset (for my own schemes) would have the PM2.5 data means for each year since MODIS data starts by pixel or whatever, and then split them into counties, provinces, etc. Looking forward to that follow-up.

  • Pingback: Tweets that mention Follow-up: Just how does China’s air quality compare globally? | Angel Hsu --

  • Alex Wang

    Great post. Very interesting stuff.

  • Donna

    Before the tsunami and the earthquakes they were the most clean country in this matter, people ove there very often lived over 100 years old

  • Anjali Makam

    people in china are poor and have no jobs.the envoirnment is bad.can’t they just make the poor people tell everyone about the issue and paid

  • Deejayhd2

    The G0v put too much growth on GDP and industry and then they do not care about the people and the environment. Now the bad thing happen. They dare not include the PM2.5 because otherwise the data will be very ‘难看’ and then the people will become angry…