Module 1 Paper:

An Analysis of Air Quality Monitoring

Systems For More Informed Policy
Making
MPI Edward Yin

10/9/2024

Acknowledgments:

ChatGPT for brainstorming and proofreading

Google Scholar for research

Abstract
In an increasingly industrialized and urbanized world the health risks associated with various air
pollutants have become more evident, leading to the necessity to monitor air quality and create effective
and targeted legislation to reduce pollution. Alongside these growing concerns, geospatial
cyberinfrastructure has also grown and is capable of monitoring air quality in real time with high spatial
 1

resolution. This paper aims to assess the strengths and weaknesses of various technologies utilized
throughout the world for air quality monitoring, such as the cost-benefit, ability to provide data in real
time, coverage, and spatial resolution in urban areas. Low-cost sensors provide high spatial resolution
data, affordable pricing, good coverage, but often face issues with calibration and reliability ranging
wildly from different manufacturers. Mobile or portable sensors also provide high spatial resolution data
in real time, however lacking the temporal resolution without adequate density. Satellite imagery provides
large coverage, but may not be able to provide data in real time or adequate spatial resolution. As the
demand for precise and accessible air quality data continues to rise, the integration of diverse monitoring
technologies into a cohesive geospatial cyberinfrastructure is essential for advancing urban air quality
management and public health initiatives.

Keywords
- Geospatial Cyberinfrastructure
- Public Informatics
- Urban Analytics
- Air Quality Monitoring
- Low-Cost Sensors

List of Tables
Tables 3 and 4 from Buehler et al.
Table 1 from Castell et al.

Introduction
As urbanization and industrialization propagates, the volume of harmful pollutants has increased, leading
to a variety of health concerns that affect the respiratory, nervous, cardiovascular system and digestive
systems. These health issues can range from a simple throat or nose irritation to debilitating
memory disturbances (Kampa et al.). Effects like these pose a significant risk to young children
and the elderly. The pollutants most commonly associated with these health complications are carbon
monoxide (CO), nitrogen oxides (NOx), sulfur dioxide (SO2), ozone (O3), and particulate matter ranging
from sizes of 2.5µm to 10µm(PM 2.5, PM10). Given the serious implications for public health, it is
imperative to develop cost-effective solutions that provide accurate, real-time data to inform the creation
of effective policies aimed at mitigating air pollution and protecting public health.

To meet these ends geospatial cyberinfrastructure is a critical tool, as it merges wide arrays of data
gathering techniques with cloud computing, allowing for a near seamless stream of information to be
funneled to the requisite parties. For the purposes of data collection, satellite imagery, low-cost sensors,
air quality monitoring stations, and drone technology have been used to collect data, each providing
unique benefits with distinct tradeoffs. The advent of low-cost sensors provides cities with unprecedented
spatial resolution allowing for per neighborhood analysis, while satellite imagery maintains a broader
scope allowing for the tracking of pollutants from areas potentially outside of the city. Yet, with these
strengths, modern day monitoring solutions come at the downside of potentially inaccurate measurements
or immense costs.
 2

The goal of this paper is to analyze the strengths and weaknesses of each technique for air quality
monitoring. Paying particular attention to the cost, real-time monitoring capabilities, spatial resolution,
accuracy, and coverage. By evaluating these technologies, I hope to elucidate how these technologies can
be leveraged in tandem to support urban air quality and create informed policies and practices to maintain
and improve public health. Furthermore, this paper seeks to identify points of future research for the end
goal of healthier and less polluted cities.

Background
Policies about air quality have dated as far back as 1897 in the United States of America, as Ohio passed a
law requiring that all steam boilers in cities be constructed or modified a way in order to reduce black
smoke emissions (Stern et al.). However, it wasn’t until 2009 that geospatial cyberinfrastructure was
utilized for the purposes of monitoring air quality (Yang et al.), a good ~60 years after the concept of a
computer had been created. From there, some of the earliest methods of air quality monitoring systems
were implemented, with low-cost sensors and drone based monitoring systems implemented about a
decade after.

Literature Review
The first technology I came across was the use of low-cost sensors spread throughout a city for real time,
high spatial resolution data collection. These sensors are far cheaper than their reference instrument
counterparts, allowing for their widespread deployment throughout various urban landscapes (Buehler et
al., 2021). This is a huge advantage for air quality monitoring in urban environments as the concentration
of pollutants can vary drastically over a small change in distance, requiring a large density of data points
to establish a gradient of pollutants (Castell et al., 2017; Karagulian et al., 2019). Despite the cheap cost
to obtain these sensors, they still maintain the upside of being able to monitor all critical pollutants and
are capable of providing data in real time, aiding in the early detection of harmful spikes in pollutants
(Buehler et al., 2021). Yet, with their ease of implementation low-cost sensors often have accuracy issues
compared to their reference counterparts. Due to this limitation, not all low-cost sensors can be used for
informing legislative bodies for public health policy decision making (Castell et al., 2017). Though the
issue of accuracy can be circumvented by calibrating each sensor with a reference tool, these calibrations
can often be difficult to implement as even small environmental differences such as heat, humidity, or
interfering compounds may cause discrepancies (Buehler et al., 2021).

Satellite monitoring provides a high spatial coverage of large areas, potentially allowing for the study of
the impact of city emissions on the surrounding environment or nearby settlements. Satellites, while
expensive to initially set up, are cheaper in the long run in comparison to setting up multiple expensive
monitoring stations with reference instruments over a large area (Holloway et al., 2021). Presumably,
these satellites would stay in service over the course of decades, providing long term historical data
allowing for a large temporal scope as well. Alongside the advent of geostationary satellites, pseudo real
time monitoring could be achieved in the area in which the satellite is deployed (Holloway et al., 2021).
However, due to how far satellites are from the surface obstructions such as clouds can prevent
monitoring, coupled with the downside that satellite imagery can only capture column concentrations of
pollutants rather than surface information can lead to discrepancies in data interpretation (Holloway et al.,
2021).
 3

Another novel method of air quality monitoring utilizes advancements made in drones or UAVs. These
methods often take two unique approaches, using either a fixed wing drone to sweep across large areas or
rotary wing drones with extreme maneuverability to access hard to reach areas. Such drones can provide
data with extremely high spatial resolution, as they can take data on intervals as small as every meter
(Jońca et al., 2022). This advantage is further extrapolated by the notion that the drone can carry a variety
of instruments and can access areas that traditional sensors may not be able to be installed in. However,
drones have a limited operation capacity, as battery life severely limits the temporal scope of operation.
The temporal scope is also further limited by unfavorable weather conditions, not to mention how the
nature of propellers can cause disturbances in the data gathered by the drones (Jońca et al., 2022).

Similar to drones, mobile air quality monitoring provides high spatial resolution data as they can be
equipped to moving bodies on the surface. These bodies can be bicycles, vehicles, or even carried by
personnel individually (Van den Bossche et al., 2015) (Wu et al., 2017). The resolution of the data can be
on intervals as small as 20 to 50 meters allowing for the visualization of pollution gradients. These mobile
sensors also enable the environmentally conscious to help contribute to the cause through wearable
sensors (Van den Bossche et al., 2015). However, a mobile approach requires that each route be run
multiple times for meaningful data collection, as during a single pass strange spikes in pollutants may
occur due to an extraneous exhaust plume (Van den Bossche et al., 2015).

Motivating Factors
The preservation of one's health has always been a powerful motivator for increasing research in any
particular topic. Air pollution is no exception to this, as pollutants are capable of acting as pro oxidants of
lipids and proteins forming free radical generators. This elevated state of oxidative stress and high
concentrations of radicals eventually leads to degenerative diseases such as atherosclerosis, heart attacks,
stoke, chronic inflammatory diseases (rheumatoid arthritis), cataract, central nervous system disorders
(Parkinson's, and Alzheimer's disease), age related disorders and finally cancer (Kampa et al.). Such
worries demand an adequate response with the stringent monitoring of the air quality in our habitable
environments for the purpose of creating well informed public policies to maintain public health.

This is particularly more important to vulnerable communities such as young children, the elderly, and
those with respiratory conditions. As such, communities with large populations of the aforementioned
groups would seek to protect and maintain the health of their loved one by demanding transparency and
strict regulations on keeping the air they breathe safe. Therefore, a robust approach to air quality
monitoring is essential for addressing these health risks and fostering healthier living environments.

Research Question or Research Objectives (150-200 words)

In order to create well informed policies governing air quality, how can a diverse system of GCI be
implemented to effectively monitor air quality in urban environments? What are some tradeoffs of each
technology utilized for data collection in terms of cost, spatial resolution, coverage, accuracy, and real
time data capabilities?

To address these questions, some research objectives have been established:

1. Evaluating performance of low-cost sensors, satellite imagery, mobile monitoring, and drones.
2. Analyzing the tradeoffs between each air quality monitoring method.
 4

Analysis
Satellite imagery offers a global perspective on air quality, making it a valuable tool for understanding
pollution patterns across different regions (Holloway et al., 2021). However, one notable drawback is its
lower temporal resolution, which can limit its effectiveness for capturing short-term air quality
fluctuations. This temporal limitation means that while satellite data can provide a broad overview of air
quality trends, it may miss critical short-term events that local monitoring can detect. However, this
drawback can be mitigated by the usage of a geostationary satellite, where the satellite stays directly
above the point of interest. This approach can allow for a better albeit not perfect temporal resolution,
where a picture can be taken and analyzed roughly every one to two hours (Holloway et al., 2021).
Although a remedy exists for temporal resolution issues, there is no fix for a cloudy day where columns
of contaminants are not observable. Satellite imagery also suffers from being unable to observe the
surface directly, creating a need for complex processing computations to make sense of the images
captured. Consequently, while satellite imagery can provide large scale coverage of a specific region, its
ability to obtain high spatial resolution data on urban areas remains tenuous at best.

Contrary to the lack of spatial resolution satellites provide, stationary air quality monitors provide
consistent and high-quality data due to their robust design and calibration (Buehler et al., 2021). These
monitors are strategically placed to capture representative air quality measurements over time. In contrast,
mobile air quality monitors are capable of collecting data across various locations, allowing for a more
dynamic assessment of air quality in ever changing urban environments (Van den Bossche et al., 2015).
This mobility enables researchers to gather information from areas that may be underrepresented by
stationary monitors (such as streets), potentially revealing localized pollution hotspots. However, to
ensure that the data from mobile air quality monitors is accurate, multiple passes of an area need to be
made. A way to implement this is by attaching mobile air quality monitors to public transportation
services such as subways, trams, light rails, or buses. This would allow for the regular monitoring of
specific routes throughout urban environments making the data cleaning process easier as outliers are
more apparent. Despite these fixes though, the issue of mobile air quality monitors remaining accurate
throughout the changing urban environment does not have an apparent fix. As these mobile air quality
monitors are typically low-cost sensors operating on a battery, they are notoriously unreliable in changing
conditions as calibrating them to all the unique situations is untenable (Castell et al.).

Low-cost sensors have gained popularity for their affordability and accessibility (Castell et al., 2017;
Karagulian et al., 2019). They can provide high spatial resolution data, particularly in regions with limited
monitoring infrastructure. However, when comparing low-cost sensors to high-end monitors, there are
concerns regarding accuracy and reliability. High-end monitors typically utilize advanced technology and
rigorous calibration processes, resulting in more precise measurements. In contrast, low-cost sensors can
suffer from issues related to calibration, which may affect the integrity of the data they collect. These
calibration and accuracy issues are severe enough for the R value between the data on site and reference
instruments to fall to below 0.5 in some cases (Castell et al.). But the calibrations of low-cost sensors can
be corrected by calibrating the sensor to the respective location where it is to be installed by use of a
referencing instrument. This process would take time to implement, but would prove to be cheaper than
installing high-end monitors throughout an entire city or neighborhood. Calibrated low-cost sensors have
demonstrated to be extremely accurate, demonstrating R values of greater than 0.88 for ozone, nitrogen
oxides, carbon monoxide and PM2.5 (Buehler et al.). While this serves as an appealing solution for
 5

accuracy issues, it cannot be understated that the lifespan and maintenance of low-cost sensors remain a
challenge that must be addressed.

Conclusion

This analysis reveals that a variety of technologies must be incorporated into a unified geospatial
cyberinfrastructure (GCI) to create well-informed public policies aimed at improving air quality, as no
single method can reliably or cost-effectively monitor an urban environment. Satellite imagery provides
general snapshots of how external pollution sources may impact a city’s air quality; however, surface-
level low-cost sensors are essential for determining pollutant concentrations at a local scale, allowing for
higher spatial resolution data. Recognizing that it is impractical to deploy numerous low-cost sensors
throughout an entire urban area, a network of mobile sensors integrated into public transportation can
effectively gather meaningful high-resolution data to identify pollution hotspots. Additionally, the need
for reference instruments and high-end air quality monitoring stations remains crucial, as both low-cost
and mobile sensors require calibration tailored to their specific locales.

In conclusion, adopting this hybrid approach not only enhances the accuracy and reliability of air quality
monitoring but also supports the development of informed policies that safeguard public health. By
leveraging the strengths of various monitoring technologies, cities can work towards creating cleaner,
healthier environments for their residents.

Directions for Future Research

Addressing the accuracy and calibration of low-cost sensors to create reliable, readily available, and
cheap sensors with universal calibration protocols throughout all sensor models for ease of use and
accessibility (Castell et al.).

Utilizing artificial intelligence and ML algorithms in air quality monitoring systems to create predictive
models. Empowered by these forecasting models, better and informed policies can be created to tackle
urban pollution problems (Karagulian et al., 2019).

References

Buehler, Colby, et al. “Stationary and portable multipollutant monitors for high-spatiotemporal-
resolution air quality studies including online calibration.” Atmospheric Measurement Techniques,
vol. 14, no. 2, 9 Feb. 2021, pp. 995–1013, https://doi.org/10.5194/amt-14-995-2021.

Castell, Nuria, et al. “Can commercial low-cost sensor platforms contribute to air quality
monitoring and exposure estimates?” Environment International, vol. 99, Feb. 2017, pp. 293–302,
https://doi.org/10.1016/j.envint.2016.12.007.

Holloway, Tracey, et al. “Satellite Monitoring for Air Quality and Health.” Annual Review of
Biomedical Data Science, Annual Reviews, 20 July 2021,
www.annualreviews.org/content/journals/10.1146/annurev-biodatasci-110920-093120.
 6

Jońca, Justyna, et al. “Drone-assisted monitoring of atmospheric pollution—a comprehensive

review.” Sustainability, vol. 14, no. 18, 14 Sept. 2022, p. 11516,
https://doi.org/10.3390/su141811516.

Kampa, Marilena, and Elias Castanas. “Human health effects of Air Pollution.” Environmental
Pollution, vol. 151, no. 2, Jan. 2008, pp. 362–367, https://doi.org/10.1016/j.envpol.2007.06.012.

Karagulian, Federico, et al. “Review of the performance of low-cost sensors for air quality
monitoring.” Atmosphere, vol. 10, no. 9, 29 Aug. 2019, p. 506,
https://doi.org/10.3390/atmos10090506.

Morawska, Lidia, et al. “Applications of low-cost sensing technologies for air quality monitoring
and exposure assessment: How far have they gone?” Environment International, vol. 116, July
2018, pp. 286–299, https://doi.org/10.1016/j.envint.2018.04.018.

Prakash, Jai, et al. “Deployment of networked low-cost sensors and comparison to real-time
stationary monitors in New Delhi.” Journal of the Air & Waste Management Association, vol.
71, no. 11, 16 Sept. 2021, pp. 1347–1360, https://doi.org/10.1080/10962247.2021.1890276.

Stern, Arthur C., and Emeritus Professor. “History of air pollution legislation in the United States.”
Journal of the Air Pollution Control Association, vol. 32, no. 1, Jan. 1982, pp. 44–61,
https://doi.org/10.1080/00022470.1982.10465369.

Van den Bossche, Joris, et al. “Mobile monitoring for mapping spatial variation in urban air
quality: Development and validation of a methodology based on an extensive dataset.”
Atmospheric Environment, vol. 105, Mar. 2015, pp. 148–161,
https://doi.org/10.1016/j.atmosenv.2015.01.017.

Wu, Yi-Chen, et al. “Air quality monitoring using mobile microscopy and machine learning.”
Light: Science & Applications, vol. 6, no. 9, 15 Mar. 2017,
https://doi.org/10.1038/lsa.2017.46.

Yang, Chaowei, et al. “Geospatial cyberinfrastructure: Past, present and future.” Computers,
Environment and Urban Systems, vol. 34, no. 4, July 2010, pp. 264–277,
https://doi.org/10.1016/j.compenvurbsys.2010.04.001.