local vs. regional

Does the air quality improve further away from the city centre? Is the air better in Shunyi than on TianAnMen? Better under a tree in the hutong than in an appartment on the fourth ringroad? Do we breathe clean air on the Great Wall? Or are we all living in that great cloud of smog over China’s east coast, supposedly visible on satellite images? (also see this post with link to recent NASA image) A comprehensive study of air pollution using satellite images can be found here.

Let’s compare the APIs for Beijing with its neighbour cities of Tianjin (100km east), Shijiazhuang (250km south), Qinghuangdao (300km east), Datong (300km west) and Taiyuan (400km south west).

 regional API june-july 2007

Observations from this graph:

  • Beijing has the worst air pollution index most of the time
  • Beijing has a pattern of more extreme peaks, the other cities’ curves are much smoother - i assume this shows that if the conditions allow pollution to build up in Beijing (no wind, no rain - a kind of greenhouse effect) the API shoots up
  • the low pollution days seems to correlate much better than the peaks, e.g. around 28 June, 1 July, 10 July, 19 July - it is tempting to see 3 cycles of around 10 days in June
  • the individual ‘bad days’ of Beijing do not seem to correlate with the other cities

My conclusion would be that their is a regional component in the air pollution, clearly seen in the ‘good days’ probably caused by wind and/or rain, while the ‘bad days’ are more of a local phenomenon, where bad air is allowed to build up.

So it makes sense to use these nearby cities’ values when trying to find out if the authorities’ attempts to make the sky blue in Beijing have any effect.

The Beijing Environmental Protection Bureau provides daily measurements for dozens of stations spread over the Beijing area, and from those data it is clear that the API can vary widely over relatively short distances of 20-50km; for example on 25 October 2007, it went from 82 in one place to 183 in another, both in the province of Beijing. Even on good days the values vary widely, for example on 7 October 2007 low was 15, high 57. I am not exactly sure how the overall number for Beijing is calculated (as visible on the SEPA website), maybe that is explained somewhere on the site. If you find it, please let me know.