China amends air quality measures but misses key pollutant – PM 2.5

The Chinese Ministry of Environmental Protection has been drafting new proposals (see rough Google translation in English here) to amend former daily reporting of the Air Pollution Index (API), which has been used the last 20 years to communicate air quality and health hazards posed by air pollution on any given day.  In 2000 the MEP (then the State Environmental Protection Agency, or SEPA) began reporting a daily API for 42 cities; now, data for 113 cities are available from the China National Environmental Monitoring Center.

Although these new specifications are still in draft form, it’s interesting to take a look to see what the MEP is considering, particularly in light of the fact that China’s annual National People’s Congress parliamentary meetings just concluded and approved the 12th Five-Year Plan (full text in Chinese here).  As I’ve written with my colleague Deborah Seligsohn of the World Resources Institute, the new Plan includes an ambitious range of energy and environmental targets, including those for air pollutants like SO2 and for the first time nitrogen oxides (NOx).  While the Plan included a blueprint for major reduction goals for these criteria pollutants, the specifics as to how targets will be allocated and policies implemented still remain to be developed over the coming months by the individual Ministries, provinces, and cities.

These draft AQI guidelines provide some insight as to how monitoring of criteria air pollutants might change as a result of the Plan.  I’ve taken a look at the second draft, which the MEP posted on their website during the last week of February, prior to the release of the 12th FYP.  Most notably, the new specifications appear to reflect a greater attempt by the Chinese MEP to make the former API more consistent with the United States’ Air Quality Index. This effort is reflected through:

  • Renaming the API the “Air Quality Index” to be consistent with the U.S.’s nomenclature.
  • Providing a consistent color classification system identical to the U.S.’s AQI color scheme.
  • Descriptions of health effects of AQI scores in language similar to that used by the AQI
  • Inclusion of new particulates carbon monoxide (CO) and Ozone (O3), which were previously absent from the API
  • Changes to the calculation methodology to reflect the U.S.’s AQI

I’ll spend the rest of this post highlighting some of these proposed changes.

Consistent Communication

Figure 1. Color classifications and descriptions of the new AQI, compared to previous versions and the US. Sources: MEP, 2011 and Andrews, 2009.

Even though the API was originally based on the United States’ AQI, there are differences in the scale and corresponding health hazard categorizations as well as color classification schemes.  Inconsistencies in color classifications within China are due to the fact that unlike in the United States, where the AQI colors are standardized, local environment protection bureaus (EPBs) in China have been allowed to set their own color schemes.  As Figure 1 depicts, the MEP is proposing a color coding scheme that is entirely consistent with the U.S.’s AQI, unlike the example Andrews (2009) shows of conflicting colors between Beijing and Guangzhou that could be confusing for travelers between the two cities.  Making the colors classifications the same as the United States will also provide more transparency and clarity for those familiar with the U.S. AQI system.

Further, while the descriptions of the AQI classes (i.e. ‘Excellent,’ ‘Good,’ etc.) haven’t changed from the previous API, the descriptions of the Health Effects are similar to those provided by the U.S. EPA.

How is the new AQI calculated?

Remember that the API was determined from only using three pollutants:  SO2, NO2, and PM10. The concentration of each pollutant is measured at various monitoring stations throughout a city over a 24-hour period (noon to noon). The average daily concentration of each pollutant is then converted to a normalized index, which means that each pollutant is given its own API score.  The daily API then only reflects the pollutant with the highest API (see Vance Wagner’s concise explanation on how the API is calculated).

Figure 2. Concentration normalization table for pollutants in the AQI. Source: MEP, 2011. 1) If 1 and 2 are the same, then use 1/2 of 2; 2) Use the concentration limit of Class 2 (TBD); 3) if the concentration of O3 exceeds 0.800 mg/m3 then it exceeds the scale.

As shown in Figure 2, the AQI now includes three additional measures: carbon monoxide (CO), Ozone (O3) – 1 hour average and Ozone (O3) – 8 hour average. The concentrations of each of these six pollutants are normalized according to the table in Figure 2.  The (1),(2), and (3) annotations mean that the MEP is still debating what concentration levels should be. Several options under consideration are found in the “Instruction Manual” document of the draft proposals (in Chinese only).

What is notably missing, however, is a measure of PM 2.5 (air particulates with a diameter of 2.5 microns or less; known to have serious health implications such as asthma, lung cancer, and cardiovascular disease, due to their ability to penetrate human lungs).  Despite the fact that many major countries report PM 2.5 concentration data, some are viewing the lack of PM 2.5 from the new AQI as a major disappointment.

Ma Jun, Director of the Institute of Public and Environmental Affairs (IPE), the Beijing-based NGO who released the Air Quality Transparency Indexwrote about earlier this year, said in a Global Times article that leaving it out was a mistake.

He goes on to say:

“Technology for measuring PM2.5 is not a problem for China any more, as cities in developing countries like New Delhi and Mexico City have already made the index public,” said Ma. He said a reluctance to include the crucial index has to do with concerns about local economies.

“Government agencies feel the index may hurt the image of many cities that want to attract investment or that they may not be able to improve PM2.5 pollution in a short time,” Ma said.

However, contrary to Ma Jun’s assessment that it’s not a technological capacity issue, I wonder if the decision to leave PM 2.5 out of the new AQI isn’t really due to local availability of PM2.5 monitoring technology.  There is a sentence (6.1 地方各级环境保护行政主管部门可根据当地的实际情况和环境保护工作的需要,参照本 标准的要求,增加空气污染物评价项目,如细颗粒物(PM2.5 等)) on Page 4 of the draft proposal that says that local EPBs can consider projects that increase the range of pollutant monitoring, specifically mentioning PM2.5.

On the other hand, one notable improvement in the AQI’s calculation is the change in methodology proposed.  While the U.S.’s AQI is based on the highest reading in a city and thus represents the “worst” air quality case a person could encounter, the Chinese API represents an average.  The draft proposals improve upon the API’s methodology, adopting a similar calculation method to that of the U.S.

First, “individual AQIs” are calculated as follows:

Figure 3. Formula for determining the Individual Air Quality Index (IAQI). Source: MEP, 2011.


IAQIp is the individual AQI;
Cp is the concentration of the six pollutants (SO2, NO2, PM10, CO, O3-1hr and O3-8 hour averages. If a city has more than one monitoring station, the average of the pollutant concentrations are used [对于城市区域为多测点的日均浓度值]).

BP(hi) is the pollutant with the highest concentration
BP(lo) is the pollutant with the lowest concentration
IAQI(hi) is the index score of BP(hi) (on the IAQI 0-500)
IAQI(lo) is the index score of BP(lo) (on the IAQI 0-500)

The max of these IAQIs (Figure 4) is then used to determine the AQI.

Figure 4. Formula to determine the AQI. Source: MEP, 2011.

I will spend some more time going through the longer instruction guidelines for the proposals and will update this post if I find more details. In the meantime, I welcome any comments or alternative interpretations.


Andrews, S.Q. 2009. “Seeing through the Smog: Understanding the Limits of Chinese Air Pollution Reporting.” China Environment Forum, Vol. 10.

Ministry of Environmental Protection. 2011. Technical Regulation on Ambient Air Quality Index Daily Report. Second Draft. Available here:

Thanks to Chris Haagen for providing some translation assistance.

Facebook Twitter Email

, , , , , ,