Among all the recent PM2.5 news, this chart from an early January CCTV report caught my eye:

I couldn’t find this data on the Beijing EPB website to verify, so if anyone out there does find an original source please let me know. It is a little strange that numbers for 2000 and 2005 are reported, since the Beijing EPB claims to have only monitored PM2.5 since 2006, but let’s ignore that for a moment and assume that CCTV has reported this accurately. This would be the first time that I’ve seen official PM2.5 numbers reported by the Beijing EPB, so they are worth examining. Specifically, let’s look at two questions:

1) Is this declining trend correct?
2) Are these absolute numbers credible?

Let’s look at these questions one at a time. The Beijing EPB claim of declining PM2.5 concentration especially surprised me because, just two months ago, Xinhua cited a Chinese Academy of Science report suggesting that PM2.5 has been increasing 3-4% per year. A more detailed description in the People’s Daily said this trend has been occurring since 1998. I haven’t seen the original CAS report, so I don’t know the details behind the study (e.g. how many monitoring stations, locations, etc.). Nonetheless, it’s not very encouraging to see this contradiction — and one coming from a government source. Let’s see how and if Beijing EPB tries to reconcile this contradiction over the year as more and more PM2.5 data come out.

As for other data sources, I still haven’t seen any public, peer-reviewed studies that show clear long-term PM2.5 trends for Beijing. If anyone out there knows of any, please let me know by e-mail or the comments. My hunch is that the few research teams out there (e.g. universities) who do have continuous PM2.5 monitoring data over a long time series are unwilling to publish them because those data are fairly sensitive, but I may be wrong.

In the absence of clear long-term trend studies, all we have are a few snapshots in the literature. Many reports just cover one or two monitors over a limited time. Still, let’s take a look at the reports as a way of gauging the credibility of the Beijing EPB-reported data. In other words, how do the Beijing EPB-reported PM2.5 numbers compare against the few other independent PM2.5 estimates we have?

Well, the first obvious place to look is the US Embassy’s monitor. The Twitter feed from the US Embassy’s PM2.5 monitor only reports hourly and daily values, although both the new foggybeijing site and Steve Andrews’ chinadialogue piece both present annual summaries suggesting that the annual average PM2.5 from that monitor was around 100 ug/m3 in 2010-2011. This is significantly higher than the Beijing EPB 2010 figures of 70-80 ug/m3. So, we have one dissenting data point, but remember it’s only one point. The Beijing EPB could easily (and perhaps justifiably) attribute this discrepancy to the fact that their data is an average over the entire city, whereas the Embassy monitor is at a single spot in a heavily-trafficked area. We’ll have to wait and see how they explain that.

Next let’s look at some reported values in the peer-reviewed literature. Here’s a good summary “lit review”-style paragraph from Chan and Yao, 2008:

PM2.5 measurements have been widely reported in Beijing in the last 10 years. He et al. (2001) made the first comprehensive PM2.5 measurements in Beijing and reported that the annual average of PM2.5 was 115 ug/m3 at Chegongzhung (an urban site) from September 1999 to September 2000 [Vance edit: the He et al. paper also reports a second site at 127 ug/m3]. The annual average was 96.5 ug/m3 at the same site from August 2001 to September 2002 and the decrease was attributed to the air pollution control measures carried out in Beijing since 1998 (Duan et al., 2006). On the basis of measurements on selected days in four seasons, Zheng M. et al. (2005) and Wang J. et al. (2004) reported annual averages of PM2.5 in 2000 and in 2001 of 101.4 mg m3 and 109.6 ug/m3, respectively. Wang Y. et al. (2005a) reported that the average of PM2.5 from 2001–2003 was 154.3 ug/m3, significantly higher than the values reported by the other studies mentioned above.

One more I found in addition to this was Yang et al. (2011), who reported PM.5 values around 2005-2006 of 118.5 ± 40.6 ug/m3.

So, we can summarize our brief data investigation as follows:

I must emphasize very clearly that all these data are not directly comparable; I present them here merely as a first-order snapshot of the PM2.5 levels at various times and places around the city. (To give a sense of how variable the levels can be across the city, consider this quote from Yang et al., 2001: “In Beijing, annual average PM2.5 concentrations varied by 50 ug/m3 (near three quarters of mean concentration at the rural site) between the paired rural/urban sites over a distance of 70km.”) But it does seem as though many of these reported PM2.5 levels are higher than what the Beijing EPB has claimed (via CCTV), and that no clear trends are visible.

Conclusions:
A (brief) review of independent literature studies does not immediately support the Beijing EPB’s reported PM2.5 numbers and declining trend for Beijing. In addition to several individual data points showing higher PM2.5 levels, it’s concerning that there is a Chinese Academy of Sciences study that apparently shows an increasing trend of PM2.5 — the exact opposite of that claimed by the Beijing EPB. However, with such limited data, we can’t really say comprehensively or definitively whether the Beijing EPB data are valid or not. The only thing I can say 100% conclusively is that I’m eagerly awaiting the release of more data that back up their claims (or, alternatively, the public release of more clear long-term PM2.5 trend studies from independent research groups).

On January 6th, the Beijing EPB made a major announcement regarding the plan for monitoring and reporting PM2.5 this year. This is huge and terrific news, with most media sites characterizing the announcement as a clear and concrete result of the recent public outcry over Beijing’s air quality (for example, see BBC, Reuters, China Daily, AP, NYT, CNN, and two WSJ pieces: “Beijing Caves…” and “Beijing Bows…“).

Because this is such a big announcement, I’m going to translate the entire thing here. I wanted to do so because there are a lot of positive developments in the announcement that weren’t mentioned in most of the brief news stories, and also because I’ve noticed some minor mischaracterizations in the media about what it actually says.

The great news, only some of which was reported by the media (not by error, probably just because of space constraints) includes:

- In addition to PM2.5 reporting, hourly ozone reporting should also start by the end of the year;
- Individual, hourly data points for all stations will be reported as opposed to city-wide averages;
- The language used to describe the pollution will be changed (presumably to reflect the fact that “Blue Sky Days” here frequently aren’t, and China’s current designation of “slight pollution” is, well, not exactly slight.)

These are all enumerated in the final paragraph of the announcement, which reads like an awesomely affirmative response to a Beijing air quality information disclosure wish list.

My only minor complaint with the international coverage is that some of it implies that Beijing will begin widespread PM2.5 reporting by Spring Festival (a couple of weeks from now). Actually, the announcement only states that some “research-type” PM2.5 data will begin to be released by Spring Festival (to give credit, the BBC was the only media I read that really nailed this). Full PM2.5 reporting for the city will (hopefully) be completed by the end of the year. The announcement is filled with all sorts of caveats about how much work it will take, and about the need for the new national standards and relevant regulations to be released, etc. In other words, Beijing is very much downplaying the expectations here. Regardless, it’s still just wonderful news that has been a long time coming.

Anyway, onto the translation:

—
Source: http://www.bjepb.gov.cn/portal0/tab189/info8308.htm
Translated by www.livefrombeijing.com
Bold added by translator

我局就启动PM2.5监测、完善空气质量信息发布制度答记者问
Beijing EPB responds to journalists’ questions about launching PM2.5 monitoring and improving the system for air quality information disclosure

1月5日，我局环境监测处负责人就本市启动PM2.5监测、完善空气质量信息发布制度回答了记者的提问。
On January 5th, a representative from the Beijing EPB Environmental Monitoring Division answered journalists’ questions on Beijing’s plans to launch PM2.5 monitoring and improve the system for air quality information disclosure.

1、关于北京空气质量监测现状
1. Regarding the status of Beijing’s air quality monitoring

北京市在1984年完成了空气质量监测系统的初期建设并投入运行，当时有8个空气质量自动监测子站，主要分布在当时的城八区。2000年以后，北京开始逐步扩建完善空气质量监测系统，至奥运会前，建成了27个环境空气质量自动监测子站，分布在全市各个区县，开展例行空气质量自动监测。按照国家关于空气质量监测的有关规范要求，监测空气中的二氧化硫（SO2）、二氧化氮（NO2）、可吸入颗粒物（PM10）等，24小时连续自动监测。并于1998年开始周报，1999年开始发布空气质量日报信息，2001开始发布预报信息。
In 1984, Beijing completed the initial phase of construction of an air quality monitoring system, and began operation. At that time there were 8 automatic air quality monitoring stations, mainly distributed in the 8 districts of the city at that time. Beginning in 2000, Beijing began expanding and improving the air quality monitoring system. By the time of the Olympics, Beijing had set up 27 automatic air quality monitoring stations spread out over all the districts and counties in the entire city, and had begun automatic monitoring. According to the requirements of the national ambient air quality monitoring regulations, concentrations of SO2, NO2, PM10, and other pollutants were automatically monitored 24 hours per day. Weekly reports, daily reports, and daily forecasts of air quality started in 1998, 1999, and 2001 respectively.

2、关于北京开展PM2.5监测的情况
2. Regarding the status of developing PM2.5 monitoring in Beijing

目前，北京市还没有全面系统的开展PM2.5常规监测。根据我市大气污染防治进程，北京市环境保护监测中心自2006年起，已经开始利用综合观测实验室对PM2.5进行研究性监测。奥运后，选择部分环境空气质量监测子站，在不同时段开展了阶段性的PM2.5研究性监测。
Currently, Beijing has not comprehensively and systematically developed regular monitoring of PM2.5. In accordance with Beijing’s atmospheric pollution prevention program, since 2006 the Beijing Municipal Environmental Monitoring Center began conducting research-type monitoring of PM2.5 using an integrated observation lab. After the Olympics, we selected a few atmospheric monitoring stations to develop staged research-type PM2.5 monitoring at different times.

3、关于PM2.5监测和公布进度安排
3. Regarding plans for monitoring and reporting PM2.5

现在我们已经启动了PM2.5监测网络建设，计划年底前全部完成。根据建设进度，完成一个点站建设就发布一个点站监测信息，同时完善整体空气质量信息发布方式。计划在春节前，首先通过监测中心网站等实时发布各监测子站二氧化硫（SO2）、二氧化氮（NO2）、可吸入颗粒物（PM10）3项常规污染物每小时的浓度数据。同时公布综合观测实验室的PM2.5 研究性监测数据，供市民参考。国家新标准和相关监测规范发布后，将按照监测规范，利用现有仪器设备先在6个监测子站开展PM2.5监测，同时发布实时数据；根据监测设备采购、调试工作进展，会逐步增加PM2.5监测子站，力争年底前完成全市的PM2.5监测站点建设并发布实时监测数据。
We have already begun building the PM2.5 monitoring network, and plan to complete it by the end of the year. In accordance with the construction progress, as we complete each station, we will report that station’s monitoring information. In parallel, we will improve the overall method of reporting air quality information. First, before Spring Festival, we plan to report real-time, hourly concentration data for three pollutants – SO2, NO2, and PM10 -  through the website of the environmental monitoring center. At the same time, we will report the research-type monitoring data of PM2.5 from the integrated observation lab for the public to consult. After the national standard and relevant monitoring regulations have been issued, in accordance with the regulations, we will use existing PM2.5 monitoring equipment to begin monitoring PM2.5 at 6 monitoring stations, and simultaneously report real-time data. In accordance with progress on equipment procurement and adjustment, we will gradually increase the PM2.5 monitoring stations, trying very hard to finish the entire city’s monitoring stations and issuing their real-time monitoring data by the end of the year.

4、关于开展PM2.5监测的准备工作
4. Regarding preparation work for developing PM2.5 monitoring

启动PM2.5监测要从四个方面做准备。一是申请资金，订购监测设备；二是优化监测网络布局、建设站点、单机和系统调试；三是升级市环保监测中心的网站，调整空气质量监测数据收集分析信息系统；四是进行人员培训。
There are four aspects we need to prepare in order to launch PM2.5 monitoring. One, apply for funding and procure the monitoring equipment; two, optimize the placement of the monitoring network, construct the stations, computers, and debug the system; three, upgrade the website of the Beijing Municipal Environmental Monitoring Center and adjust the the information system for collecting and analyzing the air quality monitoring data; four, conduct staff training.

5、关于完善空气质量信息发布制度
5. Regarding improving the system for air quality information disclosure

准备从5个方面完善空气质量信息发布方式。按照国家新标准和监测技术规范，一是发布纳入常规自动监测范围的各项污染物监测信息，包括PM2.5和臭氧的监测数据；二是借鉴国际通行做法，将过去侧重发布全市一个平均数据，改为发布各个监测子站的数据，使市民能够了解到自己生活区域的空气质量情况；三是由每天只发布一个24小时平均值改为发布各监测子站各项污染物的每小时的数据；四是以市环保监测中心网站为依托，增设空气质量信息专业发布平台；五是改进空气质量信息发布的提示语言，使之更贴近市民生活，更好为市民服务。
We are preparing to improve the method of air quality information disclosure from five aspects. In accordance with the new national standard and technical regulations for monitoring: one, disclose monitoring information from all pollutants obtained by the regular automatic monitor, including PM2.5 and ozone monitoring data; two, learning from international methods, change from the past method of strongly emphasizing the disclosure of one average data point for the city to disclosing the monitoring information from each station, in order that citizens can understand the air quality situation in the area in which they live; three, change from issuing only a daily 24-hour average to issuing hourly data for each pollutant for each station; four, entrust the Municipal Environmental Monitoring Center website to add a special platform for air quality information; fifth, change the notification language for air quality information disclosure in order to be closer to citizens’ lives and serve them better.

At times over the past few months, I have feared that the US Embassy might cave in to diplomatic pressure and close down their BeijingAir Twitter feed. At first I was worried about the recently wikileaking of a diplomatic cable describing Chinese officials’ concern that the Embassy data were “insulting” and causing “undesirable ‘social consequences.’” Then, more recently, I worried that the public tone used by Chinese media to defend Beijing’s air quality was become more aggressive (the suggestion to wear masks is “unnecessary scaremongering“; the Embassy’s “way of releasing (the data) is more like hype). I wondered if someone higher up in the US government might pull the plug.

Well, it seems my fear was unfounded. It looks like the Embassy is doing the opposite from backing down and laying low – they are doubling down and taking steps to publicize their data even more! Today I received the “American Citizen Bulletin” from the Embassy. The second news item, right after the announcement about holiday closures, was:

FOLLOW THE EMBASSY’S AIR QUALITY MONITOR

The Embassy has an air quality monitor <http://beijing.usembassy-china.org.cn/070109air.html> to measure PM 2.5 particulates on the Embassy compound as an indication of the air quality in Beijing. Although citywide analysis cannot be done on data from a lone machine, this monitor is a resource for the health of the community. It is updated hourly.

Sure enough, the BeijingAir Twitter data are now visible, inside the Great Firewall, right on the Embassy’s website.

However, in what may be at least a slight nod to diplomacy, I did notice that only the English version of the site has the data on it; the Chinese version has the explanation translated directly, but the column where the data should be is blank:

Snapshot of the Embassy’s English-language air quality site:


And the oh-so-gently harmonized Chinese version:

Every once in a while, the world’s spotlight focuses on Beijing’s air pollution, turning up the heat on the low simmer of anger, frustration, and confusion over the air quality here and threatening to make it boil over. In 2008, the spotlight came because of the approaching Olympics. In 2009, it was the launching of the US Embassy’s BeijingAir Twitter feed. In 2010, it was crazy bad air.

And now we have 2011′s version. The spotlight is back on Beijing’s poor air quality this week, not so much because of a singular event but because of a combination of a few things: a worse-than-average October, a few recent multiple-day hazardous air streaks, and a growing awareness of the disparity between the Chinese and US governments’ air quality metrics (awareness that is now spreading even more rapidly through Weibo, China’s Twitter). The Beijing EPB appears to have been caught flat-footed in responding to the recent surge of media reports – both domestic and international – that have appeared over the last few days.

But before we get into that, let’s look more closely at exactly what happened. The root cause seems to be October’s worse-than-average air quality. The Beijing’s EPB’s own monthly air quality reports for October (see 2011, 2010, 2009) reveal that the number of days exceeding China’s ambient air quality standard was much higher in October 2011 (12 days) than either October 2010 or October 2009 (6 and 7 days, respectively). Using data downloaded from MEP’s datacenter and converting to pollutant concentration (explanation here), I calculated that the average ambient PM10 concentration in October 2011 was 138 ug/m^3, which is around 16-17% worse than October 2010 (119 ug/m^3) or October 2009 (117 ug/m^3).

So we had a bad October, although let’s keep some objective perspective here: all of the numbers we’re talking about – 117, 119, 138 ug/m^3 – represent terrible air quality, and all exceed China’s own ambient air quality standard (which is already much looser than international standards). In other words, we went from terrible air to slightly more terrible air. But was the change significant enough to justify the huge increase in attention we’re seeing?

Well, confounding the situation was a series of consecutive really, really terrible days (October 8-10, 20-23, and 30-31), which the Beijing EPB attributed to poor meteorological conditions not favorable for pollution dispersion. During these streaks, the US Embassy’s meter routinely stated that that Beijing’s air was hazardous, and yet data from China’s Ministry of Environmental Protection frequently claimed just “slight pollution.” People were understandably confused – and angry.

Add to this the final ingredients: some terrific international reporting summarizing the issues (e.g. Barbara Demick in the LA Times), rapid information and rumor sharing on Weibo, and even domestic media outlets reporting on increased hospital visits and canceled flights, and the discrepancy between Chinese and US government readings. Even the Global Times featured a scorching opinion piece titled “Air pollution exposes public distrust“:

Public debates over China’s air pollution are getting intense…

Figures from the US Embassy, measured by the standard of PM2.5, revealed that Beijing’s air pollution reached “hazardous” levels. Their data is followed by more people. In comparison, Beijing authorities, who use the standard PM10, have been rebuked for deliberately hiding the truth from the public regarding air pollution.

All of this is putting the Beijing EPB on the hotseat in a major way. And the Beijing EPB – through its primary spokesperson, Du Shaozhong – has really struggled to come up with a consistent messaging strategy. They have varied between a few different types of responses:

- It’s not that bad. (October 12th in the Global Times -”It was only slight pollution, which was not bad enough for people to wear masks.“)
- The US Embassy data might not be correct. (November 2nd in the AFP – “I’m not clear about their way and methods of monitoring or how they ensure the accuracy.”
- Anyway, it’s inappropriate for China to be held to foreign standards. (November 1st in Caixin – “China’s air quality should not be judged from data released by foreign embassies in Beijing.”
- OK, we admit that we don’t monitor correctly, but we’re working on it. (November 3rd in the AFP – “The Beijing bureau applies the current national standard, which is undergoing an amendment…Technically we are ready to adopt the PM2.5 standard.”

The Chinese public isn’t buying the obfuscations. The Global Times editorial, while conceding that it is inappropriate for China at this stage of development to be held to international standards, concludes that the fundamental problem is that the Chinese government is losing credibility, because it is is afraid to tell people the truth, even though the truth is “bad news.”

James Fallows calls the current air pollution debate flare-up a “really big problem for China.” I agree, but let’s also remember that this isn’t a new problem. The key question now is, will this flare-up boil over long enough to force some fundamental changes in how air quality is measured and reported? The Beijing EPB successfully deflected the spotlights of 2008, 2009, and 2010 without any substantial changes. It will be interesting to see if this one passes with only more vague promises of future changes, or if we will finally see some concrete progress.