Remapping Rondo

The goal of this storymap is to highlight how the construction of a national highway route disproportionately affected African American communities. Historical maps and data from the Minnesota History Museum and the University of Minnesota will provide insight into how policymakers chose routes that deliberately ran through Black communities to disrupt their growing wealth and maintain segregation at a time when federal courts were striking down the traditional tools of racial segregation (Archer 2020).

This project aims to show that the construction of the I94 highway in St. Paul, Minnesota, detrimentally impacted the historically African American community of the Rondo neighborhood. Through the use of historical data on population and housing, this project will also demonstrate how neighborhood grading systems that the highway construction was based on relied heavily on racial demographics and not on housing costs, making it unreliable for highway officials to use.

In 1956, President Dwight D. Eisenhower signed the Federal-Aid Highway Act, which aimed to significantly expand the previous highway routes. Over 1.85 billion dollars was allocated to this project. A portion of this money was used to construct the I-94 highway, which now runs from Michigan to Montana.

The Highway Act anticipated roads built quickly, and it expected that over 41,000 miles of interstates would be constructed by 1969. Early construction was fast to allow for the "ambitious scale and timeline... planners faced few constraints and little opposition.” Negative quality-of-life effects, such as land taking, noise, and pollution, were not considered until construction was underway and could not be undone. (Brinkman 2020, 81-82)

While the government had given a lot of money to policymakers and engineers in charge of the project, they had not given a lot of direction. These officials “had complete control over freeway route locations.” In many cities, "building a freeway through the black community was not only the most common choice but the choice that generally had the support of the dominant white community." (Mohl 2004, 675)

The boards in charge of zoning decisions "mirror the power arrangements of the dominant society. Generally, planning and zoning commissions are not known for their racial and ethnic diversity" (Bullard 1993, 23). It was easy for the discriminatory processes that had previously shaped the United States to continue as the African American communities lacked representation.

The zoning choices that were made in the 1940s and 50s can still affect those living in modern times. Once a highway, sewage plant, or landfill is built by a house, the value can plummet, making it difficult to sell and keep a family there for generations. This can also affect the long-term health of the residents, who are exposed to higher pollution and stress levels.

Building highways in areas of predominantly African American populations is an example of environmental racism. This phenomenon refers to the unequal exposure of communities of color to environmental hazards.

An example of this is the 1979 case of Bean vs. Southwestern Waste. It stated that "from the early 1920s to the late 1970s, all of the city-owned municipal landfills and six of its eight garbage incinerators in this area of Houston were located in African-American neighborhoods." Within Houston at the time, African Americans made up only 28 percent of the population, but "82 percent of the municipal landfill sites were located in African American neighborhoods" (Bullard 1993, 24). Even though Bean eventually lost the case, it was still a landmark in environmental racism and a step toward recognizing how past zoning issues can affect populations decades later.

These hazardous facilities can also affect the health of populations that live close to them. As they are mostly built near African American communities, these groups will experience health issues at much higher rates than their White counterparts. In a study conducted by the Agency for Toxic Substances Disease Registry in 1988, it was found that "African American children are two to three times more likely than their white counterparts to suffer from lead poisoning" (Bullard 1993, 23). If a family is already unable to pay to move away from the hazardous area, they are unlikely to be able to pay for medical treatment for their children.

Seen below are three examples of African American communities that have been affected by the construction of highways. They are located in distant parts of the United States, showing that this is a widespread issue across the entire country.

Besides being the location for hazardous facilities and highways, African American neighborhoods also consistently have less green space. In 1935, the Home Owners' Loan Corporation (HOLC) completed a map that was used to show real-estate investors the most desirable areas to build. The areas were rated from A to D based on housing cost and race of people living in the neighborhood.

In Figure 5, the Original HOLC map can be seen. The areas rated "D" were the redlined sections. The practice of Redlining is "categorizing entire neighborhoods to determine the level of mortgage-lending and insurance-providing to be offered. Historically, African American residents in neighborhoods... and African American homeowners have been categorized as risks to lenders, who are therefore less likely to do business in those neighborhoods" (Ryan 2018, 538). The effects of this map can still be seen today through the differences in green space in modern neighborhoods.

The map in Figure 6 is the tree cover in the sections labeled "D" in the HOLC map. They can be seen to have a majority of red areas, depicting no tree cover at all. The greenest areas on the map, with the most tree cover, are near the river and are now parks. The map in Figure 2 is tree cover in the sections labeled "A" in the HOLC map. The percent tree cover in the greenlined districts is ~18%, while in the redlined districts, it is only ~6% (Vick 2023). Green space can remove pollution, lower noise, and cool temperatures. Environmental problems and health issues can become present in neighborhoods without urban forests (Escobedo 2011).

Research has found that "exposure to small plantings, parks, and views of urban forests may be effective in improving human health" (Pataki 2021). Health issues such as depression and anxiety can be reduced when residents have access to public parks to relax in. Trees also help filter out pollution. In St. Paul, all major highways go through these redlined districts. This causes higher pollution rates, and with almost no trees to help filter the air, residents are exposed to high levels of pollutants. This can lead to long-term health issues, such as respiratory diseases.

Environmental racism and unfair highway zoning are issues that affect all areas of the United States. One piece of this puzzle is the Rondo neighborhood in St. Paul, Minnesota. Further studying this smaller community over a period of many years can highlight generational issues that can occur by splitting a populace in half. The actions that the Rondo neighborhood takes in modern times to stay connected to their roots and promote growth in their community can be an example for other groups that face the same issues.

The Rondo community began in 1858 when Joseph Rondeau bought forty acres of land in Minnesota. Over the next ten to twenty years, he sold much of this to different developers. The neighborhood was bordered by University Av. on the north and Summit Av. to the south. In the 1930s, the community of Rondo held fifty percent of Saint Paul’s African American residents, and by 1950, this had grown to over eighty-five percent (Reconnecting Rondo 2021).

Rondo was a middle-class neighborhood with grocery stores, schools, churches, and family homes. This was a tight-knit community that had been in this location for generations now. Two popular clubs were The Credjafawn Social Club and The Sterling Club. These places for relaxation and networking were founded around the 1920s to compensate for being excluded from White clubs and professional groups. Education was also very important in Rondo, and "integrated schools, such as Central High School and Maxfield Elementary School, created a relatively high level of education and literacy among minority residents" (Alam 2017).

As early as the 1930s, citizens wanted a highway to connect the growing business districts of St. Paul and Minneapolis. The St. Anthony route was chosen for this, but some people raised concerns. One of those people was George Harrold, who had concerns over the land loss and uprootment of businesses and families. He suggested the Northern route, but "ultimately, the St. Anthony Route was chosen and approved by government officials citing its efficiency" (Kleinjung 2024). When the Federal Aid Highway Act was signed in 1956, it finally gave the city the funding it needed to start the construction (Alam 2017).

In Figure 12, the Northern Route that Harrold proposed can be seen. It would have avoided cutting through any communities and still allowed for a connection between the downtown areas of the two cities.

In 1956, St. Paul began acquiring houses through eminent domain, much to the concern of the Rondo community. Originally, the Highway Department promised over 30,000 dollars to assist residents in relocating, but this money for relocation never came. Some residents continued to resist the construction, and a prominent figure in the community, Reverend George Davis, was forcibly removed from his home by police to make way for demolition crews. Construction continued, and finally, in December of 1968, the I-94 section between Minneapolis and St. Paul was unveiled (Alam 2017).

In 1991, PBS highlighted the Rondo Community in a KTCA report. It was filmed less than 25 years after the highway was unveiled and can already demonstrate its devastating impact. One quote that stood out in the PBS clip can be seen below:

The full clip can be viewed below. It is around ten minutes long but is helpful for more background information on the Rondo Community.

The I-94 highway can already be seen to have harmed the Rondo community. But, to quantify this impact, the data on population and housing must be further studied before and after the construction. This project will also focus on comparing the before demographics to the HOLC map found in Figure 5 to show that it relied heavily on racial demographics and not on housing costs, making it unreliable for highway officials to use.

The first step to understanding the impact of the highway was to locate orthomosaic photographs of St. Paul before and after the construction. The University of Minnesota John R. Borchert Map Library had detailed images of historic maps put together for a previous project. I used this to create a zoomed-in comparison of the Rondo Neighborhood in 1956 vs. 1966. These orthomosaics demonstrate how drastically the Rondo area changed in a very short amount of time. The Rondo neighborhood is highlighted in a red outline.

Below, the St. Paul HOLC Map from 1935 can be seen again. Previously in this story map it was utilized to discuss green space in the relined zones of the city. Now, this map will be used to compare population and housing demographics.

The HOLC also included descriptions for each section, with reasoning for why the grade was given. The Rondo Neighborhood was labeled D4, and the area description read:

Officials used this HOLC data to help determine the location of I-94, and adding it on top of the 1956 orthomosaic can help people understand why the Rondo area was chosen for the highway, as it was labeled as "D." The original HOLC data layer was cut down to only St. Paul, as information on Minneapolis is not needed for this project. Rondo is highlighted with a darker red color and a black dotted outline. As this grading system was based on race and not just income and housing value, it presents issues as being used as a source by highway officials. A highway layer has also been added in dark purple to show where I-94 is located more clearly. All the highways in St. Paul can be seen to go through a neighborhood graded either C or D.

The original HOLC shape fill was published by the University of Minnesota and Minnesota National Resource Atlas. To better suit this project, I changed symbology to gradient colors to better represent the grading of the neighborhoods. The green color represents the best and red the worst-rated neighborhood, with colors in between to show B and C.

To show that the HOLC grades relied on race demographics and not just housing costs, the demographic data must also be examined. This information could be found through censuses taken years before construction. As the HOLC map was made in 1935 and highway planning started in the early 1950s, the best data to look at was the 1940s, as it was a good midpoint. Census data from 1980 and ACS data from 2020 were also used to compare the after-effects of the highway for the Rondo neighborhood. All data was found through IPUMS National Historical Data.

In IPUMS, the data selector was used to choose information such as year and variables on population and housing. All variables are examined to find average housing cost and racial demographic information. Once the source tables are selected, the GIS file that matches the data, in my case, the census tract, must also be downloaded.

This was done for all three dates, so three data sets and three shape fills must be downloaded, as the census tract shapes change slightly with every new census year.

Figure 14 shows an example of the first few rows of the 1940 data table. It includes a GIS Join column that allows the numerical data to be joined with the polygon census tract in ArcGIS Pro.

In 1940, all maps and data tables must be made by hand. Figure 15 depicts the original census tract map for St. Paul. All historical census maps were found on the National Archives website. This map further qualifies the IPUMS data, as Ramsey County has exactly 76 tracts in the data table. This is the same amount of sections the map is divided into.

The previous map is also divided into smaller sections. Those who worked on the census also created 38 smaller maps of St. Paul to give a more detailed look at each neighborhood. Figure 16 shows the census tract sections that make up the Rondo Neighborhood.

These historical maps give information about the locations of houses and roads. Studying these historical maps allows us to better understand the sections in the most recent census tract and how they have changed over time with population shifts.

The data table has information but must be joined with the shape fill to make it presentable. This is done in ArcGIS Pro by downloading the two separately and then joining them through the GISJoin number in the table.

The result is a map that has demographic data along with geographical location. The Rondo Neighborhood can be seen with the red outline.

The data is now correctly matched up, but it can still be difficult to understand with letters and numbers representing data variables. The IPUMS data also gives the user a tract book when the data is downloaded to make it understandable. Figure 17 depicts an example section of the tract book for 1940. It tells you what letters match up with specific variables.

This code system can be helpful as some tables have very long names or multiple sections. Figure 18 depicts the population by race, and it can be seen that there are fifteen different variables within it. If every name was listed, it may be confusing for the researcher if a new table is being moved onto, or the whole name may not be able to be seen.

While the code system is important for the person in charge of gathering important data and manipulating it for their use, it can confuse the audience if left like this for the final product. The names can later be changed through popup configuration in ArcGIS Online. The researcher can decide to use the original name or change it to something shorter and more suited to their study.

For this project, three different maps were made using ArcGIS Pro. So, three variable lists and three shape fills must be downloaded through IPUMS and uploaded to ArcGIS Pro. When gathering data, I tried to include the same or similar variables on the three lists to make the years easier to compare. The same process of joining was done for the 1980 and 2020 data as was done for 1940.

1980 was chosen for data as the 1970 census did not include average housing cost in their data. Choosing the later date also allowed for a lag for effects from the highway that was completed in 1969.

The 1940 census tract included fewer sections in the county's northern section than in future years. To make the data easier to compare, the 1980 and 2020 shape fills were cut down using the lasso selector tool to be the same size as 1940. As the main focus is the Rondo Neighborhood, these extra sections in the county's northern area are unnecessary.

Now that the maps are cleaned up to suit the project needs, they can be published to ArcGIS Online using the share feature. The result is three maps that can be later manipulated through symbology and popup configuration to display information in the results section clearly.

To understand the ratings of the HOLC, the demographics of 1940 can be compared to the 1935 HOLC shape fill. The census tract from 1940 was placed over the HOLC demographics to see where they line up and where there might be two or more sections within one census tract block. Different sections can be selected to view the demographic variables. The Rondo Neighborhood can be seen with a red outline. The legend is also displayed in the corner to make the color system of the HOLC map easier to understand.

Even though the maps do not line up perfectly, the data can still be used to study if the researchers in charge of creating the HOLC map relied more heavily on population and race demographics or housing value. Five of the census tract sections were selected for further study and can be seen below:

Below is the Black Population of Ramsey County in 1940, shown through graduated colors. The darker the color blue, the greater the population of Black people. The Rondo neighborhood can be seen outlined in the black rectangle. Outside of Rondo, only one census tract is medium blue in color and none are the darkest shade. Based on the above census data, about 80% of all Black residents of Ramsey County lived in the Rondo neighborhood in 1940.

The map in red represents the Average Housing Cost in Ramsey County in 1940. The black outline shows Rondo's location. Rondo contains sections with medium and light red color. While Rondo is by no means a wealthy area, it certainly is not a slum area like the HOLC describes it to be.

Studying the demographics of the 1940 census makes it clear that the Rondo Neighborhood was rated "D" largely due to the large Black population that lived there.

Now that the HOLC data has been compared to the data from the time period it was created, the impact of the highway on Rondo and the shifting demographics in Ramsey County can be studied.

The maps in blue below will compare the Black Population in Ramsey County over time. They were made by changing the symbology of the previous shape-fill maps discussed in the methods section.

The graph below depicts the Black Population over time. It compares the decades and the population in Ramsey County to Rondo. In the 40s, almost all the Black population could be seen living in Rondo. This began to shift in the 80s, and by modern times, the percentage that lives in Rondo is quite low compared to the total population. In 1940, ~80% of the Black Population lived in Rondo; now, this number is only ~12%. In some parts, this change can be attributed to population growth and an influx of immigrants from predominantly Black countries, such as Somalia. However, the divide the highway caused to Rondo and increased pollution and noise could also contribute to why fewer people choose to live there.

The maps in red below will compare the Housing Value in Ramsey County over time. They were made by changing the symbology of the previous shape-fill maps discussed in the methods section.

The table below depicts the Average Housing Cost in Ramsey County and Rondo throughout the decades. It was made with the data in the previous maps. I chose to include the table instead of the bar graph as inflation caused the amounts from the 1940s to barely show up in the graph. The percent at the bottom represents the average in Rondo divided by the average of Ramsey County for each decade. It is interesting how the percentage has increased over time despite the highway going in. Despite this, the average housing cost for Rondo is lower than the county average in every decade.

Table 1: Housing Cost Over Time

With a growing population, immigration, and changing housing laws, Ramsey County has changed greatly over the past 80 years. By studying the population and housing of the past and comparing it to today, researchers can predict where people may move and how demographics will shift in the future. A greater understanding of the past also allows for groups such as city planners or zoning boards to be more conscious of where they are constructing hazardous facilities or highways.

The maps created from the historical census data make it clear that the I94 highway harmed the Rondo Community and that the HOLC maps from 1935 relied heavily on locations of racial groups to determine their neighborhood grades.

The maps of the Black population over time show a great shift in where Black people chose to live. In 1940, almost all of them lived within the Rondo Community, but by the present day, this has shifted, with many living in the County's northern regions. While the population has grown greatly in Ramsey County, Figure 24 demonstrates how the overall percentage of Black people living within Rondo has shifted greatly. The percentage has gone down by over 68%.

The Housing Average had even more surprising results. The east section of Rondo, which is a dark red color in the 1980s, is interesting as the rest of the community seems to have lost wealth, with a lighter color on the west side of Rondo. Another surprising result was the Rondo housing cost compared to the whole of Ramsey per decade. The percentage can be seen to increase with time, with 2020 being 95.1% and 1940 being 77.5%. This shift could be because redlining laws were struck down in the 60s, which barred certain races from buying houses in specific communities. In modern times, Rondo is seen to be all light and medium red shades for Average Housing Cost. It is very close to the wealthiest region of Ramsey County, and I wonder if the highway was never built in this community if Rondo would not all be a medium or dark shade of red.

When comparing the Black Population to the Average Housing cost for 2020, it can also be seen that the northern areas with the highest Black Population today have the lowest housing prices. The southern areas with the highest Average Housing cost have the lowest Black Populations, with the lightest blue color shown in Map 11. Even though the demographics of where people live have shifted, we still see issues with differences in wealth. While today it may seem like a person could buy a house anywhere they like, past laws and decisions still shape generational wealth.

These results show that the highway planners relied heavily on the HOLC maps, which graded Black neighborhoods as lower. The highway being built through Rondo cut in half the African American community. Taking homes while not providing money to move cut down wealth and destroyed chances at generational wealth through owning a home.

Another researcher, Eric Myott, conducted similar research on the HOLC maps in the Twin Cities area. In his research, he found that the maps still have a great effect on where minority populations live today. He goes further into depth about the block-by-block residencies of minority populations. He has found that in recent times, "home mortgage lending patterns continue to exhibit unequal results, in which communities of color are denied loans at greater rates than whites" (Myott 2024). His research can be looked through further in his storymap below:

The information presented in this storymap is important as it allows the viewer to see quantitative data on Rondo and Ramsey County. While other projects have focused on HOLC maps and the destruction I94 caused to specific businesses or homes, this project was able to see the change over time and how these demographic factors shifted.

This data could be utilized by groups such as Reconnecting Rondo or future city planners to see how their projects can affect the surrounding populations. Looking to the past allows us to see mistakes that have been already made, and how to avoid them in order to protect minority groups or disadvantaged populations.

Reconnecting Rondo is currently working to make a land bridge over the highway. I-94 would go underground, and the bridge would span several blocks, with houses, stores, and park space. They have secured several donors and are partnering with government entities to make this project a reality.

Currently, they are in the planning and financing stage of the project and aspire to have construction from 2026-2030. The community is committed to rebuilding what was lost and hopes that this project will be a step toward rejuvenating this African American cultural center.

A land bridge would add a benefit discussed earlier in this story map: green space. Adding more trees and parks to this community could reduce stress and anxiety for citizens. It could also help filter out pollution from the highway, making Rondo a healthier area to live in. Adding businesses and homes to this area would also help increase wealth. Being one of the first land bridges in the region would attract tourism to the area as well and could become a stop-off for people traveling on the highway.

Rondo is not the only community that suffers from redlining. Other cities across the United States had HOLC maps created, including a close neighbor of St. Paul: Minneapolis. The website Mapping Inequality did a wonderful job collecting these maps and descriptions and putting them into one source.

This could be used as a source for future research as census data from other cities that have HOLC maps could be studied by decade to see how they have changed over time because these maps were created when the cities were developing.

Before the I-94 highway, Rondo was a growing middle-class African American community home to over eighty percent of the Black population in St. Paul. Because of HOLC maps based on racial prejudice, this community was split in half.

Based on census data from different decades before and after the highway was put in, we saw a change within Ramsey County and the Rondo Community. The percentage of the Black Population within Rondo compared to the entire community has reduced significantly after the highway. We can also see that in the present, the areas with the Highest Black Populations have the lowest Housing Value, which shows that past issues have not fully been solved.

Yet despite the injustices that have occurred to Rondo, the community continues to find ways to grow and have its story told. In the future, the community hopes to be reconnected through the land bridge to hopefully regain lost wealth and population.

I would like to thank my professors, Dr. Zorn, Dr. Mast, and Dr. Sun, for their help in advising me through this project and guiding me through difficult GIS aspects.

Alam, Ehsan. 2017. Rondo Neighborhood, St. Paul. Minnesota Historical Society. Available at: https://www.mnopedia.org/place/rondo-neighborhood-st-paul 

Archer, Deborah. 2020. “White Men’s Roads Through Black Men’s Homes”*: Advancing Racial Equity Through Highway Reconstruction. Vanderbilt Law Review, Vol. 73, No. 5, pg. 1259-1330

Brinkman, Jeffrey and Lin, Jeffrey. 2020. Early Interstate Policy and Its Effects on Central Cities. Cityscape, Vol. 22, No. 2, Two Essays on Unequal Growth in Housing: pg. 81-86

Bullard, Robert. 1993. The Threat of Environmental Racism. Natural Resources & Environment, Vol. 7, No. 3, Facility Siting, pg. 23-26, 55-56

Escobedo, Francisco, Timm Kroeger, and John Wagner. 2011. Urban forests and pollution mitigation: Analyzing ecosystem services and disservices. Environmental Pollution Volume 159, Issues 8–9, Pages 2078-2087

Klienjung, Jennifer. Last Updated 2024. Rondo Neighborhood & I-94: Overview. Gale Family Library: Minnesota Historical Society. Available at: https://libguides.mnhs.org/rondo/ov

KTCA Reports. 1991. Rondo. Public Broadcasting Service. Available at: https://www.pbs.org/video/rondo-15618/

Mohl, Raymond. 2004. STOP THE ROAD Freeway Revolts in American Cities. JOURNAL OF URBAN HISTORY, Vol. 30 No. 5: pg. 674-706.

Myott, Eric. 2024. Redlining in the Twin Cities in 1934: 1960s and 2018. University of Minnesota. Available at: https://storymaps.arcgis.com/stories/c2721b6ea0124689a9197d26fdfd6890

Pataki, Diane, et. al. 2021. The Benefits and Limits of Urban Tree Planting for Environmental and Human Health. Perspective Article, Front. Ecol. Evol., Sec. Urban Ecology, Vol. 9. Available at: https://www.frontiersin.org/journals/ecology-and-evolution/articles/10.3389/fevo.2021.603757/full 

ReConnect Rondo. 2021. History - ReConnect Rondo. Available at: https://reconnectrondo.com/vision/history

Ryan, Helen. 2018. Research Report: Were the HOLC Residential Security Maps the Origin of Redlining?. Oregon Historical Quarterly, Vol. 119, No. 4, pg. 538-543.

Susaneck, Adam Paul. 2024. “Segregation by Design.” TU Delft Centre for the Just City. Available at: https://www.segregationbydesign.com/ 

Vick, Hallie. 2023. Redlining, racial covenants, and health: Using GIS to analyze the effects of housing segregation in the Twin Cities. University of St. Thomas. Available at: https://storymaps.arcgis.com/stories/8f35e5bb4bf649408147582f40398909