Chicagoans living near heavy-traffic corridors breathe the most polluted air

When Irma Morales moved to Little Village nearly three decades ago, she vividly remembers the thin layer of dust blanketing the ground. The single mother of five lived about a mile from a coal plant.

"When I walked outside, my shoes would be covered with dust," Morales said in Spanish.

Morales joined the 12-year community-led effort to close the Crawford Power Plant.

How does Chicago's air quality affect you? Share your story.

"We shut them down," said Morales, adding she was diagnosed with a brain tumor during the campaign. "But for what? So they can bring more diesel trucks?"

The plant closed in 2012 and was replaced by a 1 million-square-foot Target warehouse bringing an estimated hundreds of trucks per day to the neighborhood. Morales and other protesters tried to stop the development.

Even the building process polluted the neighborhood. A botched implosion of a 378-foot smokestack from the old coal plant left her neighborhood blanketed in dust in April 2020.

"Why are you selling ... [our health] to the highest bidder?" Morales asked of city officials, saying her neighborhood is basically a "sacrifice zone" for industry.

Indeed, in one of the most wide-scale surveys of air quality in Chicago, some stretches of this mostly Mexican community were found to have the highest pollution levels in the city, along with portions of Austin, Englewood, Auburn Gresham, Irving Park and Avondale that see heavy traffic or are near industrial areas, an analysis of readings from newly-installed air sensors show.

The data is supplied by Microsoft, which consulted with the city and community groups before installing 115 of the sensors mostly on CTA bus shelters last summer, and has been collecting readings from them every five minutes over the past 10 months.

Even with more than 100 sensors, it's not nearly enough to cover the entire city and that inhibits a complete analysis of pollution for large swaths of the Southeast and Far South sides — areas long known to have poor air quality. Still, the data provide some of the most extensive hyperlocal measurements of air quality in Chicago, specifically in the high-pollution months of July through October 2021.

This story is part of a monthslong reporting collaboration on Chicago's air quality by the Sun-Times, WBEZ and MuckRock.

The newsrooms' analysis found that the worst air pollution levels among areas with sensors were recorded last year near the following locations:

• along 26th Street near Central Park and California avenues, and near the intersection of California and Cermak in Little Village;

• at two locations along streets near the Kennedy Expressway in Irving Park and Avondale;

• at six locations in Austin, with the highest near a bus stop at Chicago and Cicero avenues;

• and along Halsted near 74th Street in Englewood and 87th in Auburn Gresham.

Residents who live in these areas are exposed to high levels of air pollution known as particulate matter 2.5 — tiny material made up of many chemicals and other contaminants — that can lodge deep in the lungs and cause serious health problems. The pollutants measure 2.5 micrometers in diameter or smaller, which make them a fraction of the diameter of a human hair.

Fine particulate matter is the single largest environmental contributor to death and is linked to a number of health problems, including heart and lung disease and asthma. An estimated 5% of all premature deaths in Chicago can be attributed to particle pollution. Health experts compare the harm of breathing these particle pollutants to cigarette smoking.

Tens of thousands of diesel trucks travel in and out of the city's industrial corridors on a daily basis, transportation experts say, and they are a known source of particle pollution.

Chicago's poor air quality, and many of its hotspots, have been well-known to residents for decades. But the new data reveal the gulfs between neighborhoods. For example, levels of pollution in these hotspots are consistently higher than in neighborhoods that abut the lakefront on the North Side, including Lincoln Park, Lake View, Uptown and Rogers Park. The Microsoft data also provides other insights into air quality in Chicago.

The worst time of day for particle pollution readings, at least in Little Village, was the 6 p.m. evening rush hour — although 6 a.m. was also bad.

But even days when most people aren't working can be hazardous. The Fourth of July and other holidays — when many people clog the roads to vacation, attend barbecues or other events — ranked among the most heavily polluted days of the year.

But it was Friday, July 23, 2021, a 90-degree day, with wildfire smoke plumes arriving from the West, that stood out as one of the year's worst days for particulate matter. Wildfire smoke, hot temperatures and stagnant air worsen air pollution, including both ozone and particulate matter.

To be sure, the sensors are not as precise as the high-quality monitors used by the Environmental Protection Agency to take readings and set policy.

And there may be blind spots in the data on the Far South and heavily industrial Southeast sides, where the EPA monitors almost 250 facilities for air pollution.

Microsoft said in a statement that it was limited by a lack of bus shelters and gaps in cell tower connectivity that prevented devices on the Southeast Side from reporting readings. In a statement, the company said it is working on improving antenna placements.

Chicago's air is improving but still harmful to the most vulnerable

Particulate matter pollution has dropped over the past 20 years, largely thanks to government efforts, including programs under the federal Clean Air Act, to reduce harmful emissions from trucks, cars and power plants.

The levels in Chicago currently do not exceed standards set by the EPA, which maintains an annual average limit of 12 micrograms per cubic meter of fine particle pollution. But critics say that standard is way too high and the agency is expected to lower that maximum concentration this year. The World Health Organization says the level should be no higher than 5 micrograms per cubic meter, though health experts say no level of particulate matter is safe.

"Even a little particulate matter in the air is going to trigger asthma attacks and heart attacks, put people with lung and heart disease and other chronic diseases in the hospital and cause premature deaths," said Brian Urbaszewski, director of environmental health policy at the Respiratory Health Association in Chicago. It also can contribute to birth complications, developmental and mental health issues and cognitive decline.

Chicago's particle pollution is in addition to the smog, or ozone pollution, we breathe, resulting in among the worst levels of pollution in the country, the American Lung Association said last month.

Critics have said EPA standards and methodology for collecting particle pollution data are flawed. The large spaces between sensors — there are just a dozen official sensors in Cook County — provide limited data. The agency also uses computer modeling to assess pollution levels in the city where there are no sensors.

"If you rely on EPA monitors, you will miss quite a few hotspots," said Yang Liu, department chair of Environment Health at Emory University.

The more sensors, the better

In a statement, the EPA said the Microsoft and other lower-grade monitors give a wider picture of air-quality in the area although they don't meet "stringent regulatory criteria."

Chicago officials say they want to increase the number of air monitors in the city as part of a wider environmental impact study.

"The goal ultimately is to get more real-time, more hyperlocal data to help us make decisions," said Megan Cunningham, managing deputy commissioner of the Chicago Department of Public Health.

The city also is studying seven Southwest industrial corridors and the impact of truck traffic on safety and air pollution. While city economic development officials have promoted growth of transportation, distribution and logistics facilities in the area, the study is intended to address community concerns.

Traffic is a known "source of air pollution in Chicago," the city health department said in a statement. "We are doing our part to provide guidance and health-based recommendations to other departments."

"People who live or work in urban areas ... have health issues"

Even though the sensors were largely on bus stops, pollution from a CTA bus alone isn't enough to cause the higher readings in the hotspots around the city. Microsoft said it placed the sensors in certain areas to account for both population and traffic.

Still, heavy traffic is a potential source for the high readings in Avondale, where the sensor is located directly above the Kennedy Expressway on Addison Street, and in Irving Park, where the sensor is just west on a busy stretch of Irving Park Road. On average, more than 190,000 cars and trucks a day clog the expressway as it runs through the area.

The intersection of 87th and Halsted is heavily traveled with tens of thousands of cars, trucks and other vehicles driving through daily. The intersection is near two school bus depots where older, diesel-fuel vehicles contribute to air pollution. It's also near a Metra train station.

Farther north at 74th and Halsted, there is less traffic but there are railroad tracks for freight trains nearby. That sensor was particularly high for the last two weeks of July and the first week of August.

In Austin, an air sensor on the bus shelter at West Chicago and North Cicero avenues is exposed to heavy traffic at the intersection of two major streets. Cicero is also U.S. Highway 50. The area is surrounded by light industry and a large construction project nearby.

Once a thriving community, the stretch of busy roads lacks trees and greenspace.

"Most people think those kind of [pollution] issues left when the vast majority of industry left," said the Rev. Joseph Kyles, who has been pastor at The Promise Church of Chicago for 17 years.

Kyles, who grew up in Austin but now lives in Oak Park, didn't know there was an air monitor at the bus shelter outside the church's front door. He wasn't surprised Austin ranked among the most polluted spots in the city.

"It's a known problem," said the 61-year-old Kyles, who had a double-lung transplant six years ago. "People who live or work in urban areas are going to have health issues."

Still, he notes that with the economic and social challenges and a lack of adequate health care in the Austin community, "It's just one more thing."

Though the sensors were placed within city limits, they could reveal implications for surrounding suburbs, towns and villages.

One example is Cicero, where more than 90% of the population is Hispanic and the air is impacted by similar industrial and diesel pollution to neighboring Little Village. Microsoft did not install any sensors in Cicero, but there are several on 26th Street just east of Cicero Avenue, which separates Cicero and Little Village, and several above Cicero in Austin on Laramie and Central avenues. Using surrounding sensors like these, the newsrooms' analysis suggests that high concentrations of particulate matter in Little Village are similarly high in Cicero.

Back in Little Village, Morales says no one is listening to them. She's been organizing for decades to improve air quality.

"It feels like they don't care about our community even though we are homeowners and pay taxes," she said.

Morales, who co-founded the community organization Únete La Villita, told neighbors last month during a vigil near the site of the botched implosion that it's important to transform outrage and frustration into activism.

"That anger is the fuel keeping some of us in the struggle," Morales said. "It keeps us fighting for justice, accountability. We wanted and want a fair transition plan from the coal plant the community fought years to close down."

Other community organizers agree it's beyond time for government officials to act.

Tiffany Werner, a community science organizer with the Environmental Law and Policy Center who has been a liaison between Microsoft and local community groups, urged the city to begin crafting policy changes, such as limiting the number of new warehouse facilities in a given area.

They say while the new data isn't perfect, there is no need to wait for additional studies about the city's hotspots.

"The information is already out there," said Alfredo Romo, executive director of the McKinley Park group Neighbors for Environmental Justice. "It has gotten to the point — what is the city going to do with that information?"

Indi Khera of WBEZ and April Alonso and Irene Romulo of the Cicero Independiente contributed reporting.

Dillon Bergin and Smarth Gupta are reporters for MuckRock, a nonprofit collaborative newsroom that works on editorial projects with partners, and brings together journalists, researchers and the public on its news platform. Support for this project also came from Columbia University's Brown Institute for Media Innovation, which provides grants to encourage and support new technological endeavors in media.

María Inés Zamudio is a reporter for WBEZ's Race, Class and Communities desk. Follow her @mizamudio.

Brett Chase's reporting for the Sun-Times on the environment and public health is made possible by a grant from The Chicago Community Trust.

Editors: Matt Kiefer/WBEZ, Dave Newbart/Chicago Sun-Times, Derek Kravitz/MuckRock

Data Editor: Derek Kravitz/MuckRock

Data Visualization: Charmaine Runes/WBEZ

Digital Production: Libby Berry/WBEZ

What Microsoft did for its air quality pilot project

Microsoft's UrbanAir API, which maps its air quality sensors and readings, can be found here.

Chicago has several EPA regulatory stations, considered the gold standard of air quality testing; an array of cheaper sensor networks across the city and suburbs like PurpleAir and Microsoft's pilot project; and new satellite imagery that can detect a host of particulate matter, gaseous and ozone levels.

But previous low-cost sensor projects in Chicago failed to get enough accurate readings to draw firm conclusions, experts say. The new Microsoft data goes a step further.

Between April and July 2021, 115 sensors were installed across Chicago.

The first sensors were placed at EPA regulatory stations in May. An additional 100 sensors were installed in July, 80 of which were placed atop city bus shelters based on a random sampling of Chicago-area traffic and density data.

Twenty sensors were placed in coordination with environmental justice groups. Microsoft is partnered with the Environmental Law and Policy Center in Chicago, which assembled a group of community organizations to weigh in on some sensor locations including Blacks In Green in West Woodlawn, BUILD Inc. in Austin representing the border West Side, Neighbors for Environmental Justice in McKinley Park, Latinos Progresando in Marshall Square and the Little Village Environmental Justice Organization.

Microsoft's leaf-shaped air quality sensors are solar-powered, cellular-connected devices that collect readings on fine particulate matter, PM2.5, humidity and barometric pressure.

They have air vents on the top and bottom and feed data using a connection to a Microsoft cloud service into a real-time website maintained by the tech company. The devices collect data on particulate matter using a Sensirion SPS30 sensor, which is among the best performing low-cost air quality sensors in lab and field tests.

But there are limitations.

The difference in data quality between the low-cost Microsoft sensors and regulatory-grade EPA monitors is large, said Zac Adelman, executive director of the Lake Michigan Air Directors Consortium. Microsoft's calibrated +/- 29% error rate is below the EPA Air Sensor guidebook's recommended threshold of +/- 30% but the large error rate of the raw particulate matter readings makes them difficult to rely on for hourly or daily figures by sensor.

And there have been a handful of problems during the study: Some devices weren't able to get the cellular signal they needed. In the winter, devices — particularly those near tall buildings that block out sunlight — went into hibernation mode, resulting in lost data. In one instance, a sensor was stolen. One of the bus shelters where a sensor was located was destroyed in a car accident.

The temperature readings from the devices are largely inaccurate as the sensors overheated during hot days. The gaseous readings still need calibration before they can be used for analysis and thus were not released with the rest of the data.

But when the particulate matter readings are "calibrated" — compared to other sensors in the same network and EPA monitors for consistency — and compared to each other, they are revealing.

"What low-cost sensors tend to be good for is looking at patterns across time. Maybe the exact number isn't right, but you're seeing spikes or dips in the readings across time, or areas of higher or lower concentrations across an urban area — that's what I think we're seeing is really interesting with these sensors," Adelman said.

The sensors can also be used to fill in gaps where there aren't high quality expensive sensors, to identify spikes and hotspots, and to empower people with concerns about the air pollution in their neighborhoods, experts say. Microsoft said it is also working to create "a suite of analytics — including anomaly detection methods — to better enable broader accessibility to the data."

What we did to analyze Microsoft's air quality data

Microsoft gave us access to its Eclipse sensor data through the project's Application Programming Interface, or API. The API acts as a pipeline to the 115 sensors around Chicago and the information those sensors have recorded for 10 months — every five minutes, every hour of the day.

If all this data were placed into one spreadsheet, it would be more than 8 million rows.

Because of the massive amount of data, and the details in each reading, MuckRock and Columbia University's Brown Institute for Media Innovation, WBEZ and the Chicago Sun-Times aimed not to answer all questions at once, but one critical question first:

In each month since the sensors have been installed, what areas of Chicago have been PM2.5 pollution hotspots?

To do this, reporters stored the dataset in our own database, and then averaged sensor readings by hour, day and month. At each aggregation step, a so-called "data completion criteria" was applied to match the data quality standards laid out by the EPA for Hotspot Identification and Characterization. Any sensor's aggregated reading which had a sampling rate of less than 75% in the defined aggregation period was excluded. We also looked for hyperlocal correlations of these sensors and removed any sensor that had anomalous readings.

At this point, the collected readings still only showed the amount of PM2.5 recorded at each individual sensor.

To estimate how the reading of an individual sensor might explain pollution in specific neighborhoods, census tracts or community areas, reporters used a mathematical formula called Inverse Distance Weighting, or IDW. This is a popular interpolation technique, which makes use of the readings from all sensors at a given time, and uses those readings, based on the sensor's distance, to estimate the level of PM2.5 pollution in regions between sensors.

The reporters used this technique to estimate air quality levels at more than 50,000 uniformly distributed geocoordinates in the entire Chicago region and then generated tract level aggregated estimates. These estimates were then normalized based on the average estimated air quality in Chicago.

While IDW is a popular approach for modeling the spread of pollution in neighborhoods, estimates are not as accurate in regions with few nearby sensors. Microsoft's research team also cautioned that interpolations outside the edges of the sample, like those in Cicero, will be more uncertain because the network was designed for the city of Chicago.

Earlier this month, a project called ChiVes, housed at the Healthy Regions and Policies Lab at the University of Chicago, began visualizing weekly readings by sensor, census tract, and time of week. ChiVes has also collected dozens of sources to help scientists and community members explore environmental metrics alongside demographic, health and ecological data.