1. Home ›
  2. Capstone Projects ›
  3. Resolution Solution?: Assessing the added value of citizen science networks in urban air pollution monitorization

Resolution Solution?: Assessing the added value of citizen science networks in urban air pollution monitorization

Student
Kathryn Thel
College(s)
College of Science
Faculty Advisor
Stefano Castruccio
Class Year
2022

Abstract

Air pollution has been long known to pose significant threats to environmental and human health and researchers continue to uncover its specific dangers. As such, it is becoming increasingly necessary to monitor levels of air pollution exposure. Citizen science networks offer a cost-effective means of monitoring air pollution across wide spatial ranges and with great precision. However, this is only possible if the sensors implemented are of sufficient quality. This study aimed to validate the PurpleAir network of PM2.5 concentrations 1 using the EPA air quality monitors as reference data 2. From the PurpleAir data, high-resolution maps of air pollution exposure in New York City were produced using spatial interpolation and human
exposure was assessed using census tract data. Achieving more precise and widespread coverage of air pollution monitoring is essential in order to identify possible causes of increased air pollution, inform policy, and address potential instances of environmental injustice in the form of disproportionate distribution of environmental hazards.

Introduction

Given the significant threats posed by air pollution to environmental and public health, it is imperative that accurate monitoring measures are implemented and maintained and that their data is readily accessible to citizens. At present, Environmental Protection Agency (EPA) stations serve as the most prominent method of monitoring air pollution in the United States. However, their coverage for individual cities is very sparse. As an example, in 2020, there were at most eight active air quality monitors installed throughout the five boroughs of New York City, the city of interest in this study. Simulations serve as an alternative analytical tool used to study air pollution levels, but they can be prohibitively expensive and require substantial expertise to run. Satellites can also be used, but do not measure air pollution directly and have a spatial resolution insufficient to monitor air quality at the urban level. Raising justice and equity concerns, health modeling analyses informing regulatory decision-making and policy research are at present often conducted at spatial resolutions too coarse for close examination of the relationships between air pollution exposure and population health susceptibility, allowing inequalities within cities to remain undetected (Kheirbek, et al., 2016)3. Citizen science, the public collaboration in scientific research among nonprofessionals, offers a cheaper alternative to the EPA’s expensive sensors as a means of monitoring exposure to air pollution. PurpleAir is the most popular and widespread air pollution monitoring network and uses sensors that collectively produce a map of real-time PM1.0, PM2.5, and PM10 measurements, reporting mass concentrations every 120 seconds4. In comparison to the EPA’s eight monitors in New York City operated in 2020, PurpleAir’s map reports observations from approximately forty active sensors across the city. While more accessible and less expensive (PurpleAir’s PA-II air quality sensors cost $249.00), the sensors are not maintained by professional scientists and rely on optical rather than weight measurements of particulate matter, which is considerably cheaper but also less precise5. In the past 10 years, the cost of air pollution monitors has been lowered significantly, allowing their installation in homes and offices, for example. The higher density of sensors distributed across the country enables the creation of maps of higher resolution. The question of interest in this study is the added value of PurpleAir air pollution monitors individually installed throughout New York City, in comparison to the EPA’s AirData air quality monitors. PurpleAir data was validated by referencing daily mean PM2.5 concentrations from EPA monitors active in New York City in 2020 and performing a linear regression model. Maps were produced from the PurpleAir data using R software and spatial statistical interpolation, providing a value for every point in New York City. In addition, population data from the 2020 Census Bureau was cross-referenced with the air quality data. A statistical analysis was performed on these maps to determine the frequency with which areas of New York City experience exposure above the EPA’s National Ambient Air Quality Standard for PM2.5 (12 μg/m3as of 2013) in 2020.

End Notes

1. PurpleAir (2021a) Sensor data download tool,
2. EPA (2021) Download Daily Data, https://www.epa.gov/outdoor-air-quality-data/download-daily-data
3. Kheirbek, I., Haney, J., Douglas, S., Ito, K., Matte, T. (2016) The contribution of motor vehicle emissions to ambient
fine particulate matter public health impacts in New York City: a health burden assessment. Environmental
Health 15(89). https://doi.org/10.1186/s12940-016-0172-6
4. PurpleAir (2021b) FAQ, https://www2.purpleair.com/community/faq
5. PurpleAir (2021b) FAQ, https://www2.purpleair.com/community/faq