Hello, Air Quality World!

Welcome to the QuantAQ blog! We founded QuantAQ in 2019 as a spinout of MIT and Aerodyne Research, Inc. We realized that the most efficient path to obtaining ubiquitous air quality monitoring was to move our research outside academia and build a team to combine modern computational data science with extensive expertise in atmospheric chemistry. Our mission is to help people measure, understand, and solve their air pollution problems.

Measuring air pollution reliably using low-cost air quality sensors is hard.

Air pollution represents the single largest global environmental risk factor for premature mortality and morbidity [1, 2]. In other words, air pollution kills people - more than 7 million people each year [1, 2, 3]. The fundamental question is, how do we 'solve' the world's air pollution problems? The first step towards more effective mitigation of air pollution is accurate, reliable, and widespread monitoring of key pollutants. Distributed low-cost air quality (AQ) sensor networks are emerging as a powerful and cost-effective tool to reveal the often invisible forces of air pollution. When used correctly, AQ sensor networks can help:

1. Identify sources of pollution that disproportionately impact the air
2. Inform the design, implementation, and assessment of policies aimed at lowering pollution levels
3. Empower people and organizations to quantify the pollutants in their air and make data-driven decisions to lower their exposure and liability

How is ambient air pollution measured?

Currently, air pollution is measured by federal and state regulatory agencies via a network of very expensive, very accurate reference monitoring stations with the goal of monitoring regional AQ and ensuring regional compliance with existing regulations. Each station is roughly the size of a small trailer (see image below) and contains approximately $150k worth of monitoring equipment. In addition, each station requires a continuous supply of calibration gases and other consumables as well as upkeep by a trained technician resulting in annual operating costs that are comparable to the capital cost of the station itself. Data from each location is often reported on an hourly, 8-hr, or 24-hr interval and can often be found online; however, these data are often difficult to access and understand. While regional changes in AQ are easily tracked over these long time intervals, locally relevant changes in AQ are often much more rapid (on a minute-by-minute time basis).

Reference monitoring stations play an important role in understanding regional AQ, but unfortunately, they result in large monitoring gaps. The prohibitively high cost of traditional monitoring equipment has prevented AQ monitoring throughout many regions of the world. Low-cost, distributed AQ sensor networks play a vital role in filling in both local and global AQ monitoring gaps.

Cost-effective monitoring solutions are necessary to enable decision-makers and community members alike to understand and solve air quality problems.

How can QuantAQ contribute?

QuantAQ brings together more than a decade of experience with AQ sensor research and development. Together, we have designed, built, and deployed hundreds of sensor systems and numerous sensor networks across the world. Over the years, we have gained extensive knowledge about the strengths and limitations of these sensors. We have made mistakes, iterated on designs, and ultimately deployed improved sensor systems. We formed QuantAQ to more effectively scale and support these efforts and to bring accurate, reliable low-cost AQ sensors to market.

Very few commercially available, low-cost AQ sensors are capable of providing actionable AQ data. It is important to acknowledge that the technologies used in low-cost AQ sensors are fundamentally different than those used in the regulatory and research-grade devices. The different technologies introduce inherent uncertainty in the measurements - for example, most low-cost gas sensors (most often, electrochemical sensors) suffer from environmental degradation and interference. A large part of the work and research we have conducted has centered on developing algorithms and calibration strategies to minimize the impact these interferences have on our sensor systems. We conduct fundamental research both in-house and through extensive partnerships with academic research groups. This research directly informs our products and is published in peer-reviewed, open-access academic research journals. Being honest and transparent about the way we obtain our data is extremely important to us at QuantAQ and we will continue to support the open-science ecosystem throughout our work. Air quality sensors are inherently imperfect. Taking an honest and transparent approach to data handling is one step towards combating these imperfections.

The scale of this environmental health crisis necessitates a collaborative effort. We are much stronger standing together as a global AQ community than apart.

Where we're headed

We hope this blog will serve as a resource for those interested in learning more about air pollution, air quality sensors and monitoring, atmospheric chemistry, and the application of machine learning to understand and tackle local air pollution problems.

Please reach out with any questions at hello@quant-aq.com or follow us on Twitter at @quant_aq.

References

1. Burnett et al., Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proceedings of the National Academy of Sciences. September 2018.
2. Cohen et al. Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015. April 2017.
3. Forouzanfar et al., Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. The Lancet. October 2016.
4. Hagan, David H., Measuring ambient air quality using low-cost sensors. October 2019. Massachusetts Institute of Technology, PhD dissertation.