Air Quality Monitoring

By Priyanka deSouza

The World Health Organization (WHO) released a report in 2012 stating that an estimated 6.5 million people died (11.6% of all deaths in 2012) because of poor outdoor and indoor air quality, making it the single biggest environmental health risk on our planet. These findings have acted as an impetus for countries to better monitor air quality, and institute laws and regulations to reduce air pollution. However, globally, current monitoring efforts remain uneven. Many rapidly urbanizing countries in sub-Saharan Africa and South Asia have a dearth of air quality monitoring stations.

In order to see current ambient air quality monitoring data available, check out OpenAQ! It is an open source platform that aggregates and shares data from continuous monitoring projects all over the globe. You can go to their website to see a complete map of monitoring sites.

The OpenAQ map indicates that developed countries, such as the US and Western Europe, have many air quality monitoring stations, while there are very few in the rest of the world. This is troubling because the burden of air quality is not distributed equally, with more than 90% of deaths due to air quality occurring in middle- and low-income countries. These are precisely the countries that lack monitoring stations.














Troubles with reference air quality monitoring stations

One reason why many developing countries do not have air quality monitoring stations is cost. Reference air quality monitoring stations—stations with high-precision instruments that provide high-quality, real time, continuous air quality monitoring data—are expensive. The cost for a single station can be between USD $100,000 to $200,000! An example receipt for such a station in the US can be found here. Notably, this price tag does not include operating costs which can also be quite high. Thus reference monitoring costs can be prohibitive.

Consider a country like Nigeria, with a population of 182.5 million. Nigeria (GDP/capita is only USD $2,460/person, versus the US which is $55,836/person) currently has only one air quality monitoring station that provides public, openly accessible data, as visible on OpenAQ. This is troubling, as air quality can vary dramatically across a city, let alone a country. Although the single data point in Nigeria gives us some idea of air quality, it does not really help us understand pollution hotspots, the location of possible sources, or how much of the pollution in the country is from local, regional, or transnational sources. These insights are crucial for the development of effective air pollution management plans. However, this one data point is still useful, as it can be integrated with complex atmospheric transport models to give a better idea of air quality at a regional scale.

Cost alone cannot account for the dearth in monitoring. The costs of reducing air pollution, including monitoring, are more than compensated by the number of lives saved by these measures. Another reason for the lack of monitoring, then, is that skilled personnel are required to operate and maintain reference stations. Even if air quality data does exist, governments must have open data policies to make this data publicly available. Finally, strong institutional frameworks are required in order to interpret this data and lead to effective action.


Low-cost air quality monitors: a possible solution

In the last few years, low-cost (costing under USD $3,000) air quality monitors are on the rise. These monitors are not as accurate as the reference monitors that institutions such as the EPA use, but they are orders of magnitude less expensive. More and more communities are turning to these instruments to track air quality in their neighbourhoods. Our lab at MIT, the Senseable City Lab, works on interpreting the data from these low-cost monitors. Given the low data quality, what insights can communities obtain from this data? How can this data be leveraged to develop pollution management plans? These are important questions to help render the data communities collect meaningful.

One of the projects we worked on was examining the data from a low-cost air quality monitoring network I was responsible for deploying in six schools in Nairobi. When we analysed the data, we noted several peculiar things. In a school in the suburbs of Nairobi where we expected the air to be relatively clean, we saw huge spikes in particulate matter on Wednesday mornings. When we talked to the school, we found that the school burned waste every Wednesday morning! We thus recommended that the school either stop burning waste, or burn the waste far away from the students dormitories to prevent student exposure to harmful levels of pollutants.

Further, by examining the diurnal patterns of air pollution at each school, we noted the times when air quality peaked, and could recommend that break times in the school be adjusted so that the children were not outside when air quality was at its worst. Moreover, when we examined the air quality data with respect to wind speed and wind direction, we were able to tell where the pollution was coming from and identify potential sources at each school. You can read more about our work in our paper.

Now, in our deployment in Nairobi, we had not calibrated our low-cost monitors with high quality instruments. We therefore focused on using the trends we obtained from our monitors, instead of using the absolute values of pollutants reported. We also spoke to the community, gave them access to our data, and ran through potential sources with them in order to vet our results. We found this an effective technique, and a good way to raise awareness about air pollution.

However, scientists are developing exciting statistical techniques to use data from a co-located reference monitor to calibrate low-cost monitors. This way, corrected low-cost monitor readings can be used to report accurate absolute values of pollutants. Yet, the process of co-location and calibration can be costly and can substantially increase the cost of using low-cost monitors, and research to develop techniques to lower these calibration costs is ongoing.


Satellites: the alternate air pollution monitors

Another alternate air pollution monitoring technique is the use of satellite data. NASA can measure some pollutants from space! For example, the Terra satellite has two instruments, MISR and MODIS, that can detect particulate matter from space. These instruments measure the optical extinction of sunlight, termed as aerosol optical depth (AOD) — a measure of how much sunlight has been blocked from reaching the earth because of scattering and absorption from these particles. The AOD is a columnar index, while ground-based monitors measure surface levels of air pollution. Scientists use atmospheric models that provide the seasonal vertical distribution of aerosols in order to calculate the fraction of the AOD at the surface. They then calibrate this surface scattering coefficient with ground-based reference air quality monitors to estimate total air pollution from space.

Satellite data is exciting—it provides information about global air pollution since the late 1990s! However, the current satellites are polar, which means they only can provide snapshots of information about air pollution for a given location. The MISR instrument, for example, reports air pollution once every 2 days for places close to the poles, and once every eight days for places close to the equator. The spatial resolution of this instrument is 17.6 square kilometers. NASA is in the process of developing geostationary satellites to measure air pollution in continuous measurements of air pollution for given locations. Satellite data is the United States is currently used to detect exception air pollution events, like forest fires.


Future expansion of air quality monitoring techniques

One of the big reasons why governments such as the United States are still not using air pollution monitoring datasets fully is because the current reference air pollution monitoring paradigm prioritizes accurate air quality measurements. This is because in the US, reference air quality monitors play more of a policing role to check if air quality exceeds air pollution standards. States can accrue high penalties if air quality within their boundaries exceeds standards, and thus it is important that air pollution is measured accurately to make a comparison with the standards that can hold up in court. We thus need to expand our view of monitoring to something that can be more than a method of policing in order to accommodate the insights that we gain from using alternate monitoring techniques.

The US EPA is well aware of this issue, and has created several programs to support citizen science. It is only to be hoped that these programs continue under current EPA leadership. NASA also holds several training programs to try and make it easier for the public to access satellite data. I am thus hopeful about our collective ability to tackle the serious problem of air pollution.


About the author: Priyanka is a first year PhD student at the Department of Urban Studies and Planning where she works on issues related to air quality. Prior to coming to MIT, she worked at UNEP and was part of the team that deployed a low cost air quality monitoring network in the city of Nairobi.


Tag: Learn about the importance of air quality monitoring and the current state of monitoring around the globe. Also learn about new techniques to monitor air quality using low cost monitors and satellites