Crazy bad air

Open sharing of pollution data helps Chinese netizens brave the politics of murky air.

Crazy bad air - SOP 2013

by Natasha Vizcarra
October 11, 2013

Open sharing of pollution data helps Chinese netizens brave the politics of murky air.

It is obscure. It is scientific jargon. But to Chinese urbanites, PM2.5 is now as ubiquitous as tea. The term refers to airborne particles—from vehicle exhaust and from burning wood and coal—that are so small, several thousand could perch snugly on the pointy end of a sesame seed. These tiny particles can slip into your lungs and blood stream and aggravate heart disease and other illnesses. When PM2.5 levels in the cities rise, the chatter on Weibo, the Chinese version of Twitter, swells. Case in point: when concentrations rose to hazardous levels in Beijing on January 12, 2013, there were nearly 40 million messages on “pollution” and “PM2.5.”

“The term is technical and wonky,” said Angel Hsu, who studies Chinese environmental policy at Yale University. “But Chinese citizens have become very aware of air pollution issues, specially those involving fine particulate matter. I even saw an ad for a rock music festival called PM2.5.” Because of the extremely bad air days in Beijing, residents, visitors, scientists, and policymakers are discovering that open access to air quality data could help warn people when everyone—children, the elderly and even healthy adults—should stay indoors because of dangerous pollution. The data are crucial in the dialogue between citizens and the government in cleaning up the air for years to come. Hsu said it all began in early October 2010 when Beijing residents looked out their windows and saw not the city’s cosmopolitan skyline, but something that looked like a whole lot of wu.

Photograph of retirees performing tai chi during a smoggy day in China
Retirees perform tai chi during a smoggy day in Fuyang, China in January 2013. (Courtesy ImagineChina)

Beyond index

Wu is the Mandarin word for fog,” said Hsu. “Often, Chinese won’t describe a poor air quality day as the result of air pollution, which is called wuran. They will say that it’s fog.” However, it was not a misty blanket of wu; it was wuran so dense that locals could not make out buildings a mere hundred feet away. The thick smog caused road accidents and flight cancellations in the city, and lingered in the area for about a month. Although Twitter is blocked in China, people found a way to check air quality readings tweeted from a monitor installed at the U.S. Embassy in Beijing and were shocked at the numbers that they saw on a particularly smoggy mid-November day.

The Air Quality Index (AQI) had surged to 557, way beyond the hazardous threshold of 500. AQI levels of 301 to 500 mean that children, older adults and people with heart or lung disease should remain indoors and that everybody else should not exercise outdoors. The embassy ran out of adjectives for AQI levels beyond 500, and had assigned the phrase “Crazy Bad” to AQI levels above 500. Of course, when embassy officials realized that the Crazy Bad AQI level had actually been reached, they quickly replaced it with the more benign “Beyond Index.”

However, word of the gaffe had already spread worldwide through news reports and social media sites. Residents were confused when they compared the U.S. Embassy’s readings with the Chinese government’s numbers, which said pollution levels were moderate. While the embassy’s numbers were based on PM2.5 particles, the Chinese government’s figures were based on PM10 particles, which are larger, coarser air particles, like dust. Government-approved data on PM2.5, which have greater health impacts than PM10, were simply not available to the public. Hsu said, “Many Chinese citizens felt something wasn’t right. That’s when they turned to the Internet. They really felt that the Chinese government was hiding information from them.”

People and pollution

China is not the only country that had not been offering PM2.5 data to its citizens. Erica Zell, an environmental engineer at Battelle Memorial Institute said, “There simply aren’t enough ground-based air quality monitors in many regions of the world.” Even where governments measure air quality, they may focus on different parameters, like NOx or PM10, or the measurements may not be well calibrated. Zell’s colleague, Jill Engel-Cox, added, “Some countries do not have monitoring programs at all.”

Although ground instruments like the one in the U.S. Embassy in Beijing are useful in alerting people to their daily exposure to PM2.5 pollution, data that span several years can say a lot about air pollution trends. Governments need these data trends to make decisions about air pollution policies. Zell, Engel-Cox, and colleague Stephanie Weber knew that there were not enough ground instruments in China and worldwide to tease out these trends. “We wanted to consistently measure air quality across the globe so policy makers can better address the sources of this both global and local problem,” Engel-Cox said.

So they developed a way to use satellite and population density data to measure country-level PM2.5 all over the world for the last ten years. The researchers at Battelle collaborated with Aaron van Donkelaar and Randall Martin at Dalhousie University to make PM2.5 maps using aerosol optical depth (AOD) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Multi-Angle Imaging Spectroradiometer (MISR) sensors on the NASA Terra satellite. They also used a world population data product from NASA's Socioeconomic Data and Applications Center (SEDAC) to weigh the results by population distribution so that PM2.5 concentrations in more densely populated areas were more heavily weighted than concentrations in less populated areas. “Pollution concentrations are highest near population centers, since the industrial and transportation sectors are leading sources of emissions,” Zell said.

The researchers then worked with Hsu and her colleagues at the Center for International Earth Science Information Network (CIESIN) at Columbia University to integrate these data into the 2012 Environmental Performance Index (EPI). Hsu said, “The index compares countries on a key set of environmental issues, air quality being one of them. While developing this index for the last twelve years, we did not have country-level PM2.5 data because not all countries are at the stage of development where they can measure it,” Hsu said. “Using satellite data gave us a more consistent measure of PM2.5 at the country level, using the same data set and the same measures for each country.” Their findings showed that while China’s air pollution problems have been in the spotlight in recent years, India had the worst air quality in the world, followed by Bangladesh and Pakistan. China came in fourth.

Data map showing global annual average of human exposure to PM2.5
This map shows average annual human exposure to PM2.5 by country for 2010. Reds and oranges indicate the highest levels of exposure; India, Bangladesh, Pakistan, and China scored the worst. (Data courtesy NASA, van Donkelaar et al., 2010, processed by Battelle Memorial Institute)

Seeing through the smog

As news of the 2012 Environmental Performance Index results added to the PM2.5 chatter in social media, Chinese citizens seemed to be getting results from their Internet campaigns. In November 2011, a Chinese property tycoon launched a Weibo campaign calling for tighter air pollution monitoring procedures in the country. In two days, China’s vice minister of environment conceded that Beijing relied on a “limited” system of pollution measurement.

Chinese citizens were also sharing other sources of data. In January 2012, Chinese filmmaker Michael Zhao launched a Web site featuring daily skyline images of major Chinese cities and other major cities in the world. The site, China Air Daily, also published the U.S. Embassy PM2.5 readings, along with views of the cities from space, using MODIS images from NASA's Land, Atmosphere Near real-time Capability for EOS (LANCE) system. Zhao’s site was featured in news stories around the world and drew even more attention to China’s air quality woes.

A pair of photographs comparing clear and smoggy days in Beijing
These photographs compare Beijing in November 2011 on a clear day (left) and a smoggy day (right). (Courtesy P. Barwick)

Around this time, Hsu said, the Chinese government announced that it had been collecting PM2.5 data all along but had not seen the need to release it and that PM2.5 concentrations in China had actually decreased over the last ten years. “None of the data have been transparent,” Hsu said. “So none of the citizens believed it. That was a big lesson for the Chinese government; transparency is absolutely essential for effective environmental management and communication.”

A wind shift

Finally, in January 2013, China began publishing PM2.5 air quality ratings for seventy-four of its cities. “It looks pretty consistent with how you would expect air quality to be in these cities—a lot of them are showing very severe PM2.5 pollution,” Hsu said. China has also started to address some sources of pollution. A month after publishing PM2.5 data, the government announced a plan to curb emissions with a ban on sub-standard diesel-fueled vehicles.

Chinese citizens continue to be passionate about improving the air they breathe and compulsively check both official Chinese air quality readings and those tweeted by the U.S. Embassy in Beijing. On January 12, 2013, netizens noted that AQI readings at the U.S. Embassy in Beijing reached a staggering 755. The government did not publish the actual value, but only noted that it was “Beyond Index.”

Hsu says it is a great start. She said, “Scientists say that if you start releasing PM2.5 data today, citizens will have a better idea of the dangers and the extent of the pollution that they are facing. Still, experts in China say it will take about twenty to thirty years to clean up the air in Beijing.”


Battelle Memorial Institute and Center for International Earth Science Information Network (CIESIN)/Columbia University. 2013. Global Annual Average PM2.5 Grids from MODIS and MISR Aerosol Optical Depth (AOD). Palisades, New York USA: NASA Socioeconomic Data and Applications Center (SEDAC).

Center for International Earth Science Information Network (CIESIN)/Columbia University, International Food Policy Research Institute (IFPRI), The World Bank, and Centro Internacional de Agricultura Tropical (CIAT). 2011. Global Rural-Urban Mapping Project, Version 1 (GRUMPv1): Population Count Grid. Palisades, New York USA: NASA Socioeconomic Data and Applications Center (SEDAC).

Hsu, A., A. de Sherbinin, and H. Shi. 2012. Seeking truth from facts: the challenge of environmental indicator development in China. Environmental Development 3: 39–51.

Kahn, R., J. Martonchik, D. Diner, M. Garay, and M. Bull. 2009. MISR Level 2 Aerosol. Hampton, Virginia USA: NASA Atmospheric Science Data Center (ASDC).

Levy, R. C., L. A. Remer, D. Tanré, S. Mattoo, and Y. J. Kaufman. 2006. Updated 2009. MODIS Level 2 Aerosol. Greenbelt, Maryland USA: NASA Level 1 and Atmosphere Archive and Distribution System (LAADS) Distributed Active Archive Center (DAAC).

van Donkelaar, A., R. V. Martin, M. Brauer, R. Kahn, R. Levy, C. Verduzco, and P. J. Villeneuve. 2010. Global estimates of exposure to fine particulate matter concentrations from satellite-based aerosol optical depth. Environmental Health Perspectives 118(6): 847–588, doi:10.1289/ehp.0901623.

For more information

NASA Socioeconomic Data and Applications Center (SEDAC)

NASA Level 1 and Atmosphere Archive and Distribution System (LAADS) Distributed Active Archive Center (DAAC)

NASA Atmospheric Science Data Center (ASDC)

NASA Land, Atmosphere Near real-time Capability for EOS (LANCE)

Moderate Resolution Imaging Spectroradiometer (MODIS)

Environmental Performance Index

About the data used
Satellites Terra Terra
Sensors Moderate Resolution Imaging Spectroradiometer (MODIS) Multi-angle Imaging Spectroradiometer (MISR)
Data sets Global Rural-Urban Mapping Project, version 1 (GRUMPv1) MODIS Level 2 Aerosol MISR Level 2 Aerosol
Resolution 30 arc-seconds, latitude/longitude 10 kilometer 16 x 16 array of 1.1 kilometer radiance pixels
Parameters Human population density Aerosol optical depth Aerosol optical depth
DAACs NASA Socioeconomic Data and Applications Center (SEDAC) NASA Level 1 and Atmosphere Archive and Distribution System (LAADS) Distributed Active Archive Center (DAAC) NASA Atmospheric Science Data Center (ASDC)

The photograph in the title graphic shows fog and smog blanketing Beijing in January 2013. (Courtesy T. K. Hjorth)

Last Updated: Nov 14, 2017 at 11:21 AM EST