Nonprofit, nonpartisan journalism. Supported by readers.


The Minneapolis Foundation generously supports MinnPost's Community Sketchbook coverage. Learn why.

Huge gap in pollution exposure by race surprises U of M researchers

REUTERS/Fred Prouser
The study found that people of color are exposed to 38 percent more outdoor nitrogen dioxide, which comes from vehicle exhaust and power plants, than whites.

By now you probably know that people of color nationwide have, among other things, higher rates of unemployment, poverty and educational disparity compared to white people.

Julian Marshall
University of Minnesota
Julian Marshall

Here is one more fact to add to the list: People of color are also exposed to nearly 40 percent more polluted air than whites, according to a University of Minnesota study released Tuesday.

The study looks at the differences in pollution exposure by race, income, education and other categories throughout the country, said Julian Marshall, the study’s lead researcher and a civil-engineering associate professor at the University of Minnesota.   

“The main ones are race and income, and they both matter,” Marshall said in an interview. “In our findings, however, race matters more than income.”

The study, “National patterns in environmental injustice and inequality: Outdoor NO2 air pollution in the United States,” found that people of color are exposed to 38 percent more outdoor nitrogen dioxide (NO2), which comes from vehicle exhaust and power plants, than whites.

Nitrogen dioxide concentrations are also higher for low-income groups than for those with higher incomes. Likewise, the level of nitrogen dioxide is higher among people with less than a high school education compared to those with a high school education or above.   

To Marshall, some of the reasons that race “really matters” in the study is that when he looked at the exposure gap between high-income Hispanic and low-income white groups, nitrogen dioxide concentrations were higher among high-income Hispanics.   

“We were quite surprised to find such a large disparity between whites and nonwhites related to air pollution,” Marshall said. “Especially the fact that this difference is throughout the U.S., even in cities and states in the Midwest.”

Nitrogen dioxide concentrations in the Twin Cities, as in most cities, are higher near downtowns and areas with major highways. And Minnesota is among the top 15 states in the nation with the largest exposure gaps between people of color and whites. New York, Pennsylvania and Illinois top the list.

Video abstract: environmental justice in NO2 air pollution in the United States

Air pollution health risks

The study found that the health implications of the exposure gaps between nonwhites and whites are alarming.

For instance, the study states:

Breathing NO2 is linked to asthma symptoms and heart disease. The researchers studied NO2 levels in urban areas across the country and compared specific areas within the cities based on populations defined in the U.S. Census as “nonwhite” or “white.”

The health impacts from the difference in levels between whites and nonwhites found in the study are substantial. For example, researchers estimate that if nonwhites breathed the lower NO2 levels experienced by whites, it would prevent 7,000 deaths from heart disease alone among nonwhites each year.

The Environmental Protection Agency has added NO2 to its main list of air pollutants to monitor. The list includes ozone, carbon monoxide, sulfur oxides, nitrogen oxides and lead.  

Air pollution remains the world’s biggest environmental health risk, according to the World Health Organization (WHO). In a report published last month, WHO said 7 million people worldwide died in 2012 as a result of pollution exposure.

Ibrahim Hirsi can be reached at Follow him on Twitter at @IHirsi.

Comments (14)

  1. Submitted by Paul Udstrand on 04/16/2014 - 12:08 pm.

    Not sure why they were surprised

    Observers have made note of where poor and working class people live relative to the more affluent for centuries and it’s always been in the dirtiest and most polluted areas. I welcome the results but naivete is the explanation I can suggest for the “surprise”. Didn’t these guys see: “Slumdog Millionaire?”

  2. Submitted by Presley Martin on 04/16/2014 - 12:09 pm.


    I’m not sure why these researchers were surprised, as far as I knew non-white populations have been talking about environmental racism for years.

    • Submitted by Bryan Reed on 04/16/2014 - 01:50 pm.


      I might not be understanding what you mean by “environmental racism” but can you please elaborate? Why is it considered racist that more white people can afford to live in areas with less pollution than non-whites? Would they not be racist if they bought homes next to a factory? Also, compared to whites, non-whites tend to live in more urban areas instead of rural ones but does that mean they were forced to live there? Usually property is less expensive in rural areas so its not more financially burdensome for non-whites to live further away from highly polluted areas. It seems as if its their choice to live in the city compared to out in the country.

      But then again, I might just be misunderstanding what you mean by environmental racism…

      • Submitted by Paul Udstrand on 04/16/2014 - 03:09 pm.

        Let me explain

        It’s institutional racism, things don’t just break down along race lines because that’s the natural order, there are advantages for “Whites” built into the system. A lot of it is the result of various historical legacies that simply tilt towards a privileged skin color or cultural background.

        • Anonymous Submitted by Anonymous on 04/16/2014 - 08:26 pm.

          There is nothing systematic in place preventing urban migration.

          Alternatively, one could posit that people who are similar tend to stick together (e.g. Irish, Chinese, etc.)

          • Submitted by Paul Udstrand on 04/17/2014 - 09:44 am.

            Shuo, you just don’t see it

            I’ll give you one example, go up to North MPLS and try to find a bank. If you’re downtown or in the suburbs you have banks all over the place. You go to North MPLS and you have check cashing and pay-day loan shops. The absence of local bank branches changes the whole community dynamics, this is why certain areas of the city were hit harder by the sub-prime crises than others.

            Furthermore you’re simply missing relatively recent history. As late as the 70s realtors had unofficial “red lines” beyond which no property would be sold to people of color, and in some places one way or another those kind of restrictions are still enforced. Up until the late 40s people actually wrote enforceable property covenants that actually prohibited any future sale of property to Jews or Blacks. For a couple decades after covenants were declared unconstitutional they were nevertheless enforced via “gentlemen’s agreements”. When I talk about historical legacies this is the kind of stuff I’m referring to.

            Did you know that MPLS was once considered the most antisemitic city in the country?

            Just google: institutional racism

  3. Submitted by Eric Sandeen on 04/16/2014 - 03:55 pm.

    Income vs. Race

    So far the comments seem to be “well duh, less desirable, more polluted locations are cheaper” and using income as a proxy for race.

    But the study’s author said, “In our findings, however, race matters more than income.”

    So if it’s more about race than about income (i.e. more about skin color than house-buying power) then yes, that sounds like it involves racism somewhere along the line.

    I could imagine, just maybe, that when interstates cut through urban areas, they were more likely to be routed through poor minority neighborhoods than poor white neighborhoods, for example – although I don’t know this for a fact.

  4. Anonymous Submitted by Anonymous on 04/16/2014 - 08:24 pm.

    Non-story (or look at the impact of population density)

    This is one of those instances where you really want the actual values. Putting it at 40% and claiming “huge” is a bit hyperbolic.

    ” average NO2 concentrations are 4.6 ppb (38%, p<0.01) higher for nonwhites than for whites, 1.2 ppb (10%, p<0.01) higher for people below versus above poverty level, and 3.4 ppb (27%, p<0.01) higher for lower-income nonwhites than for higher-income whites." -- this is parts per billion.

    if you look at the data table the huge disparity comes from the urban-rural distinction. it's 14.2 ppb in urban areas vs 4.4 in rural areas. the title may as well reflect this (Urban areas 200% more polluted than rural areas!), but no one would be surprised. And we all know what the racial breakdown in urban vs. rural area is, don't we. Interestingly, the difference in pollution exposure isn't so great when we examine across socioeconomic status (1.2 ppb for above/below poverty line), but I suspect it's because poor white people in rural areas combined with poor colored people in urban areas balances out middle/upper class people in suburban areas.

    Urban vs. rural pollution exposure is completely consistent throughout the analysis independent of most variables (~15 vs. 4). Interestingly though, whites in urban areas are exposed to slightly less pollution than nonwhites. I would suspect this is because of a higher density of non-whites in larger cities (e.g. Boston, New York, Washington D.C.). Although the authors of the original article already indicate this as a weakness.

    The point here is that is that the minnpost article seems to be creating a race issue out of thin air. Whether this is intentional or not doesn't really matter, since it's obvious they're hiding the raw data and statistics, relying on a percentage to exaggerate the information. To be fair, they are quoting the study, so how much fault really lies with minnpost? Either way, in my opinion, the data and analysis is not specific enough to draw these types of conclusion.

    • Submitted by Steve Titterud on 04/17/2014 - 10:40 am.

      I scratched my head after looking at the raw data, too…

      …as the study did not appear to support the headline-grabbing use of inflammatory language such as “environmental injustice”.

      Here is another interesting point, a disconnect between the authors’ contentions and the EPA, which you will find at It shows a table of relative values of ambient nitrogen dioxide in ppb, and the range of effects of each.

      The lowest range denoted is 0-50 ppb, which the EPA typifies as:

      “No health impacts are expected when air quality is in this range.”

      Also, the same EPA document states, “Monitoring studies have shown that within approximately 50 meters of heavy traffic/freeways, NO2 concentrations may be 30 to 100 percent higher.” So there is an extreme variability depending on precise location.

      The values in the study are far below 50 ppb, yet the authors “estimate” that thousands of lives are lost in a difference of 1,2, or a few ppb in exposure “each year”.

      Maybe it’s time to demand these authors show us how EXACTLY they arrived at this estimate and why it is a reasonable estimate and not speculation or exaggeration in support of a dramatic headline !

  5. Submitted by Paul Udstrand on 04/17/2014 - 10:08 am.

    Depends on the stats

    “But the study’s author said, “In our findings, however, race matters more than income.”

    All we’re seeing here is tables, not the entire study, we really don’t what kind of stats were performed. And I note that these are engineers, not epidemiologists. Sometimes a variable’s significance will be exaggerated simply because of the statistical method, not because of it’s actual significance. I doubt that skin color actually attracts more pollution.

    In order to sort this out you’d have to look more closely at the different groups. It’s possible for instance income is showing up as a less substantial variable simple because of the cohort. Are you comparing high income Somalis to other Somali’s or to the population as a whole? We know that Somali’s, Hispanics, and Native American’s have lower median incomes than “whites” so when you say income ranks behind “race”, what exactly are you saying? Are you saying a Somali Lawyer living in Wayzata is breathing in more pollution than his neighbors or are saying that a Somali cab driver with a high income relative to other Somali’s living down on Cedar Ave. is inhaling just a much pollution as his neighbors?

    It will still break down along racial lines, but economics and policy determine where these populations are concentrated. Somali’s didn’t end up down in the Cedar Riverside area because they didn’t like Lake Minnetonka or Stillwater. Hispanic businesses aren’t concentrated along Lake Street because they didn’t like the malls of Maplewood or Ridgedale. There are built in conditions that limit their choices.

    • Submitted by Eric Sandeen on 04/17/2014 - 11:10 am.

      Sure, but…

      “I doubt that skin color actually attracts more pollution.” – not in the natural world, but like I said, it’s possible that siting decisions about pollution sources may involve the skin color of those ultimately affected.

      In other words: communities can site themselves near pollution, and pollution can be sited near communities. The former probably has more to do with income; the latter less so.

      But most of these decisions were made long ago (power plants, interstates) so even if originally they were imposed on a particular community, there has been time to shift, if they had the means. But the momentum to stay in a community can be pretty strong.

      • Submitted by Paul Udstrand on 04/17/2014 - 11:59 am.

        Not all communities “site” themselves.

        Eric, you keep assuming that economic conditions have no effect on relative freedom and that everyone has the same options. Frankly that’s a typical assumption from a privileged perspective. To this day we have a legacy of Chinatown’s in many cities in the US due the Chinese Exclusion Act that was in effect until 1943. It takes generations and decades before some policy decisions fade, if ever. Somali’s, Hmong, and Hispanic populations didn’t “site” themselves in the Twin Cities, or even in Neighborhoods within the Twin Cities, they were directed there by the US government. Nor did Jews “site” themselves in North MPLS, no in the sense that they looked around and said: “well we be anywhere but we choose to be here.” Historically the location of immigrant communities is determined by a number of predictable factors, look up Swede Hollow and Bohemian Flats… and note where they were located. We’re not talking about a white suburban migration here.

        • Submitted by Eric Sandeen on 04/17/2014 - 04:30 pm.

          I think you misunderstand me

          Paul, I think we’re in agreement here; there may be a problem with semantics.

          I do not assume that everyone has the same options. When I say “communities site themselves” that’s taking into account all the outside pressures and choices they may be forced to make; I mean they are making hard choices, not that they are deciding between unlimited, wonderful options. In some cases perhaps they had no choices; today, those choices might be primarily constrained by income, but not entirely.

          I understand that communities with less power, fewer choices and more outside pressure may end up in less desirable locations – i.e. possibly nearer existing pollution sources.

          The point I was trying to make was that once a minority community has established itself in some location for whatever reason, it may LATER run the risk of having other outside institutions choose to place new pollution sources nearby. And that decision may be influenced by institutional racism.

Leave a Reply