Scientists Have Discovered New Air Pollution Sources Using “Roving Sentinels”

Air Pollution Earth Planet

A pioneering study by University of Utah and EDF utilized Google Street View cars for detailed air quality monitoring in Salt Lake Valley. Revealing hyper-local pollution hotspots and highlighting environmental justice issues, the research marks a significant advancement in understanding and addressing urban air pollution’s uneven impact.

In the Salt Lake Valley, cars equipped with advanced air quality measurement tools, similar to Google Street View vehicles, traversed neighborhoods to gather highly detailed air quality data. This comprehensive sampling revealed distinct variations in pollution levels within different local areas. Additionally, a novel atmospheric modeling technique was developed to accurately pinpoint the sources of these pollution emissions.

In 2019, a team of atmospheric scientists from the University of Utah, in collaboration with the Environmental Defense Fund and other partners, introduced an innovative approach to air quality monitoring in the Salt Lake Valley. They equipped two Google Street View cars to function as mobile air pollution detectors, capable of identifying hyper-local pollution hotspots.

Over the following months, John Lin, a professor of atmospheric sciences at the University, developed a groundbreaking modeling technique. This method combined wind-pattern modeling and statistical analysis to trace pollutants back to their exact sources. This technique provided a level of detail in pollution tracking that surpassed the broader, less precise methods of traditional air quality monitoring, which typically assessed air quality over entire urban areas.

In a U- and Environmental Defense Fund (EFD)-led study that was recently published in the journal Atmospheric Environment, the results are in. 

“With mobile vehicles, you can literally send them anywhere that they could drive to map out pollution, including sources that are off the road that previous monitoring missed,” said Lin, who also serves as associate director of the Wilkes Center for Climate Science & Policy. “I think the roving sentinel idea would be quite doable for a lot of cities.”

Google Street View Car

A Google Street Car loaded with air quality instrumentation. Credit: Logan Mitchell

The researchers loaded the vehicles with air quality instrumentation and directed drivers to trawl through neighborhoods street by street, taking one air sample per second to create a massive dataset of air pollutant concentrations in the Salt Lake Valley from May 2019 to March 2020. The observations yielded the highest-resolution map yet of pollution hotspots at fine scales—the data captured variability within 200 meters, or about two football fields. 

“The big takeaway is that there is a lot of spatial variability of air pollution from one end of a block to another. There can be big differences in what people are breathing, and that scale is not captured by the typical regulatory monitors and the policy that the U.S. EPA uses to control air pollution,” said Tammy Thompson, senior air quality scientist for EDF and co-author of the study.

Air quality patterns were as expected, with higher pollution around traffic and industrial areas. Pollutants were higher in neighborhoods with lower average incomes and a higher percentage of Black residents, confirming a well-known issue of environmental justice. This pattern traces its legacy to redlining policies from a century ago when the Homeowner’s Loan Corp. created maps that outlined “hazardous” neighborhoods in red ink. The redlined neighborhoods often had poor air quality due to industrial activities that existed alongside residents, who were often People of Color. Urban planners exacerbated the environmental issues by using the maps as justification to build highways and permit industrial companies in the so-called hazardous areas.

“Air quality is not a new issue. It’s been around for decades and decades, and was probably much worse back then,” said Lin. “The I-15 corridor follows these redlined neighborhoods. And sadly, there’s a fair bit of research supporting the fact that redline neighborhoods, from 80 years ago, still matters. Those are in the neighborhoods still struggling with air quality problems. The legacy of racial discrimination is still there because they tend to be the under-invested neighborhoods.”

Schematic of Monitoring System

A schematic summarizing the steps in the new atmospheric model for source localization. Credit: Lin et. al (2023) Atm. Enviro

A street-level view of air pollution

The research-grade instrumentation installed in the Google Street View cars measured ambient air pumped in from the surroundings and distinguished chemical signatures of key air pollutants, including nitrous oxides (NOx) emitted by automobiles, trucks, non-road vehicles, and powerplants; black carbon (BC) from incomplete combustion from road and off-road diesel vehicles and industrial kilns; fine particulate matter (PM2.5) from dust or ash; and methane, primarily from the landfill. The researchers directed the drivers to sample air from 26 neighborhoods, from the industrialized areas of North Salt Lake to residential areas as south as Cottonwood Heights and West Jordan. The researchers chose neighborhoods that represented contrasting demographics across the valley, including in proportion of Black residents, average incomes ranging from 34K to 100K-plus, and areas dominated by industrial or residential buildings.

Most pollutants showed a strong pattern that reinforced what we already know—NOx, PM2.5, BC, and CO2 levels were elevated along highways in the valley. Areas with high levels of one pollutant were likely to be high in other pollutants, either from a single source emitting multiple pollutants or from overlapping sources.

A Case Study To Test the Atmospheric Model To Determine the PM2.5 Emission Hotspot Near a Gravel Pit Operation

A case study to test the atmospheric model to determine the PM2.5 emission hotspot near a gravel pit operation. The Google Earth image in c) shows the gravel pit—corresponding to the grid cells with the highest correlations in b). Credit: Lin et. al. (2023) Atm Enviro

“It’s kind of boring to say, ‘Well, there’s pollution on the roads.’ Everyone knows that. Right? So, we wanted to use the data to find the sources off the road,” Lin said.

The authors tested Lin’s new atmospheric modeling method with two case studies of well-known pollution sources—a large landfill methane source and a known gravel pit PM2.5 source. 

They then applied the model to analyze a previously unknown area of elevated PM2.5, located in an industrial area just south of the Salt Lake City airport. 

Next steps

The authors hope that other places will utilize the new method to identify pollution hotspot sources to make their cities safer, including identifying temporary sources, such as gas leaks, and permanent sources, such as industrial sources. Roving sentinels could help policymakers enact regulations and more effectively utilize resources to mitigate damage to their citizens. 

The authors hope to utilize the atmospheric model for projects such as Air Tracker, a first-of-its-kind web-based tool that helps users find the likely source of air pollution in their neighborhoods. Run on real-time, trusted scientific models and coupled with air pollution and weather data and developed in partnership with the U, EDF, and the CREATE Lab at Carnegie Mellon University, Air Tracker helps users learn more about the air they’re breathing, including pollution concentrations and its potential sources. Air Tracker is live in Salt Lake City Valley and will be rolled out to more locations across the country in the next couple of months.

“There are a lot of important environmental justice aspects to this work,” said Thompson of the EDF. “We need to be able to understand what average air pollution looks like in different communities, and then understand why there is variability and why there are hotspots, and therefore what we can do about it. It’s really, really important as we learn more and more about inequity in air pollution and what we’re breathing across the country.”

Reference: “Towards hyperlocal source identification of pollutants in cities by combining mobile measurements with atmospheric modeling” by John C. Lin, Ben Fasoli, Logan Mitchell, Ryan Bares, Francesca Hopkins, Tammy M. Thompson and Ramón A. Alvarez, 2 August 2023, Atmospheric Environment.
DOI: 10.1016/j.atmosenv.2023.119995

The research utilized resources of the U’s Center for High Performance Computing for computing the spatial distribution of pollution and developing the methodology for locating emission sources.

Other authors of the article are Ben Fasoli of the U’s Department of Atmospheric Sciences, Logan Mitchell of Utah Clean Energy, Ryan Bares of the Utah Department of Environmental Quality, Francesca Hopkins of the Department of Environmental Sciences at University of California, Riverside, and Ramón Alvarez of the Environmental Defense Fund.

The study was funded by the Environmental Defense Fund. 

3 Comments on "Scientists Have Discovered New Air Pollution Sources Using “Roving Sentinels”"

  1. That’s a good idea. They already pointed out they’re testing pollution on the roads, so finding the highways have pollution is to be expected in traffic. Regarding the off-road samples, they seem to be driving a 2010 Hyundai Santa Fe with a V6 gasoline-engine SUV, so their “hyperlocal pollution” measurements might be limited to the air immediately surrounding the vehicle itself. If they do this with enough passes, it could identify pollution sources when winds shift, but of course they are polluting by doing this; stationary air quality sensors would be the more logical solution, or maybe on bicycle.

    I don’t think this study should be specifically targeting the “People of Color” for how much more they were found to be polluting, but so long as they are treated equally as individuals, limit “what we can do about it” to a pollution education campaign at most. Bringing racialism into this really makes their higher “Black Carbon” uncomfortably accusatory, and they’re targeting a tiny approximately 6000-person 3% demographic of Salt Lake City. I don’t like the ominous sound of what these Californian researchers and New York activists want to “do about” them by associating Black people or any ethnicity with pollution, and I don’t think they should be imposing their racism on Utah.

    Salt Lake City has a terrible smog problem. Because it’s in a valley nearly surrounded by mountains, the weather develops what locals call “inversions”, or bubbles of trapped stagnant air. You’ll hear it only happens rarely, and only in the winter, and it’s improving, but don’t believe any of that.

  2. “The big takeaway is that there is a lot of spatial variability of air pollution from one end of a block to another.”

    Is that all spatial variability, or also temporal variability that is being measured because it takes time to drive to point B? That is a disadvantage of using cars instead of an aerial platform to obtain an instantaneous snapshot.

  3. Great tool can see a future of better understanding, I would propose that in conjunction with those devices that a development of airborne pathogens like covid/ flu etc. be included as a monitoring system. Needed at a greater scale.

Leave a comment

Email address is optional. If provided, your email will not be published or shared.