Land-use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city

Peter C Ibsen; Anna Bierbrauer; Lucila M Corro; Zachary H Ancona; Mark Drummond; Kenneth J Bagstad; Jay E Diffendorfer

doi:10.1371/journal.pone.0317988

. 2025 Mar 3;20(3):e0317988. doi: 10.1371/journal.pone.0317988

Land-use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city

Peter C Ibsen ^1,^*,^#, Anna Bierbrauer ^2,^3,^#, Lucila M Corro ¹, Zachary H Ancona ¹, Mark Drummond ¹, Kenneth J Bagstad ¹, Jay E Diffendorfer ¹

Editor: Christopher A Lepczyk⁴

PMCID: PMC11875366 PMID: 40029852

Abstract

Home Owners’ Loan Corporation (HOLC) maps illustrated patterns of segregation in United States cites in the 1930s. As the causes and drivers of demographic and land-use segregation vary over years, these maps provide an important spatial lens in determining how patterns of segregation spatially and temporally developed during the past century. Using a high-resolution land-use time series (1937-2018) of Denver, Colorado, USA, in conjunction with 80 years of U.S. Census data, we found divergent land-use and demographics patterns across HOLC categories were both pre-existent to the establishment of HOLC mapping and continued to develop over time. Over this period, areas deemed “declining” or “hazardous” had more diverse land use compared to “desirable” areas. “Desirable” areas were dominated by one land-use type (single-family residential), while single-family residential diminished in prominence in the “declining/hazardous” areas. This divergence became more established decades after HOLC mapping, with impact to racial metrics and low-income households. We found changes in these demographic patterns also occurred between 2000 and 2019, highlighting how processes like gentrification can develop from both rapid demographic and land-use changes. This study demonstrates how the legacy of urban segregation develops over decades and can simultaneously persist in some neighborhoods while providing openings for fast-paced gentrification in others.

Introduction

The recent availability of easy-to-access historical Home Owners’ Loan Corporation (HOLC) redlining data and HOLC-designated boundaries from the “Mapping Inequality” project [1] has resulted in new examinations of health, spatial, and ecological outcomes associated with the historical housing segregation mapped out by HOLC areas [2–4]. Present day land-use and land-cover characteristics such as lower tree canopy coverage, amount of green space and parks, and their associated concerns such as urban heat islands, poor air quality, and environmental pollution have all been connected to HOLC spatial extents [5,6]. However, some of this recent work has been critiqued as being either [1] too causal (i.e., HOLC maps have caused present day pattern) in their interpretations [7], or [2] reliant on limited analyses including poor consideration of patterns found in the different HOLC categories, minimal temporal analysis, and omitting narrative data found in HOLC records’ descriptions of neighborhoods [8]. Therefore, to better understand why current economical, physical, and ecological inequities exist, researchers could assess how patterns of inequality developed over time.

Until now, studying urban land-use and land-cover patterns over time has been difficult due to the lack of high-resolution historical land-use data spanning multiple decades. In this paper, we quantify the spatial and socioeconomic patterns from approximately when HOLC maps were created through the following decades using a novel U.S. Geological Survey (USGS) dataset tracking land use in the Denver metropolitan region from 1937 to 1997 [9], coupled with historical U.S. Census data obtained from the IPUMS (formally known as the Integrated Public Use Microdata Series) National Historic Geographic Information System at the University of Minnesota (www.nhgis.org) [10]. Relating multi-decadal land-use data, demographic patterns, and HOLC-designated boundaries identifies linkages between patterns developed during the periods that HOLC influenced lending practices and subsequent trajectories of investment and disinvestment within former HOLC-graded neighborhoods.

Multiple federal and local policies contribute to the patterns of racial and ethnic segregation in U.S. cities before and after the 1930s. ‘Redlining’ maps—the HOLC color-coded maps rating local mortgage markets on sociodemographic variables such as race, ethnicity, and income as well as housing stock conditions created to allow for mortgage refinancing in the wake of the Great Depression [7]—are regularly studied because they provide clear spatial boundaries in hundreds of cities. However, neighborhood racial segregation existed before the creation of HOLC in the form of racial covenants and other localized discriminatory housing practices [11,12], and there is growing consensus that HOLC policies reinforced, rather than created, inequitable access to affordable mortgage insurance [2,13]. Especially, as described in Hillier 2003, there is little evidence that HOLC maps were actively used in categorically denying mortgages to redlined areas, although discrimination could have still be reinforced at the individual homeowner-scale as local brokers took over foreclosed properties [13]. In the decades following HOLC mapping many other trends and policies influenced patterns of investment and disinvestment in U.S. cities. Urban renewal projects of the 1950s–70s resulted in large physical changes in urban centers, largely in non-White dominated districts [14,15]. From the 1950s to the 1980s urban White residents left city centers for the suburbs, while Black Americans moved from the rural south to cities during a second wave of the “Great Migration” cementing racial segregation within urban areas [16].

Therefore, present-day demographic and land-use spatial patterns in cities reflect past urban policies intended to create long-term changes [14]. When such policies built on existing patterns of segregation or were biased toward particular racial or ethnic groups, neighborhood outcomes diverged [17]. Biased policies and practices have been observed across many nations. In Germany, renters with Turkish last names and pronounced accents were less likely to find rental units [18]. Whereas, in the Netherlands, ethnic minorities applying for loans in “bad” neighborhoods were screened more rigorously [19]. Research in post-apartheid South Africa found that perceptions of a neighborhood’s value drove home loan approval rates not the personal and private wealth investments actually present [20].

Land-use and land-cover characteristics associated with inequitable housing policies such as higher multi-family residential density, lower tree canopy density, and higher urban heat vulnerability [3,12,17,21–23] also make these neighborhoods vulnerable to gentrification [24].Gentrification and displacement, the processes of higher-income and more-educated residents moving into lower-income neighborhoods, are spurned by variations in housing-stock age, vacancy rates [25], and city and neighborhood-specific factors [26–28]. As reinvestments in U.S. urban centers increase, these types of land-use and land-cover characteristics may be affected differently by changes to zoning codes, speculative real estate development, or re-greening efforts [29–32]. The spatial patterns of this connection between the built environment, past discriminatory policies, and the potential vulnerability to neighborhood gentrification have been connected from the present to the HOLC boundaries [33], but to understand how these spatial patterns emerged over time, time-series data are needed.

Our work addresses the gap in knowledge and data about how urban demographic and land-use patterns developed over time, how they were reified by redlining policies, and reinforced by continual urban renewal policies. Moreover, the diverging demographic and land-use patterns created ideal conditions for extensive changes in present day neighborhood socio-economic demographics through rapid parcel-level reinvestment leading to neighborhood gentrification. Because potentially demographic exclusionary policies still exist, or access to housing is protected through discriminatory gate-keepers in nations worldwide [19,34,35], an understanding of how these patterns occur over time, and the possible outcomes of such practices may only be noticeable long after the fact. Long-term studies can thus provide key information in understanding how urban areas subject to discriminatory housing policies develop and change their demographics and land-use over multi-decadal periods. By using HOLC maps as a starting point, our data analysis can analyze how unique patterns of land use and demographics developed and apply this analysis to anticipate present day patterns of gentrification.

Through our data synthesis, in a case study for Denver, Colorado, we address the knowledge gap of redlining effects through time, rather than examining the present-day conditions. We furthermore examine how redlining legacies influence the often interrelated patterns of land use and urban demographics. In this study, we ask two key research questions regarding the legacy of outcomes resulting from housing segregation and potentially unequal lending programs:

How have patterns in land use and demographics developed and diverged in formally categorized HOLC areas across Denver from 1937 to 2019?
How have patterns of inequality in land use and demographics from the previous decades influenced rapid present-day changes in Denver’s demographics and resulting neighborhood-specific patterns in gentrification?

Methods

Study area

Denver is located at the foot of the Rocky Mountains of Colorado, bordered by the mountains to the west and the High Plains to the east. The Denver metropolitan statistical area’s annual growth rate of 1.48% makes it one of the fastest growing cities in the western United States, and its current population is 715,000 [36]. Although Denver has a densely developed urban downtown, most of the region is characterized by pockets of medium density surrounded by large tracts of suburban and peri-urban development. Growth across the Denver region has increased steadily since the mid-1850s when White settlers first arrived in the area. Outside the city’s boundaries, most land use was agricultural—market farms, large homesteads, and small ranches—and small rural communities. A large influx of federal workers into Denver before and after World War II boosted demand for single-family housing, and the Federal Housing Administration (FHA) helped defense workers purchase homes, contributing to an already large federal presence [37]. Between 1940 and 1950, Denver’s population grew by 29%, and the city itself expanded in area from 58 to 99 square miles [38].

As the Denver metropolitan area expanded and its suburban population grew, land use and demographics changed. Neighborhoods in the periphery of Denver experienced greater upward income mobility compared to the central urban core [39]. In 1958, the Denver Urban Renewal Authority (DURA) was created, with a primary task to “eliminate post World War II slum housing conditions. [40]” This task coincided with the passage of the Federal Housing Act of 1949, which provided billions of dollars to local municipalities to redevelop areas termed “blighted” [41]. Over the following decades, Denver, and other cities across the country, entered a period of “urban renewal” and attempted to increase the property value of its urban core [14]. Into the start of the 21^st century, Denver’s population and income grew substantially [39], maintaining growth through the Great Recession (9.5% growth between 2000 and 2010), and boomed afterwards (17.7% growth between 2010 and 2020). During the same period, suburban growth was rapid but declined from the 2000s to the 2010s (24.7% between 2000 and 2010 to 15.6% growth from 2010 to 2020) [42].

Analysis of land-use diversity

To better understand present-day inequities in the United States’ cities, we examine a case study of the progressive, long-term impacts of discriminatory policies through land-use time-series data with the historical HOLC-designated boundaries in Denver Colorado. We identify the relative proportions of land-use classes within each HOLC category and track how those land-use patterns and demographics shifted over time (Fig. 1).

Fig 1 — A: Location of Home Owners’ Loan Corporation (HOLC) categories on a present-day map of Denver, Colorado. B, Land-use data, for each 20-year period (1937, 1957, 1977, 1997), from Drummond et al., 2019 [9], with outlines of each HOLC area and their corresponding identification. The legend included refers to these land-use categories. Basemap from U.S. Geological Survey, The National Map, 2024, HOLC outlines are republished from Mapping Inequality [1] under a CC BY license, with permission from Robert K. Nelson, original copyright 2023.

The USGS dataset by Drummond et al. [9] derived detailed land-use information for the years 1937, 1957, 1977, and 1997 in Colorado’s Front Range. The result is a 1-meter resolution land-use map, with a minimum mapping unit of approximately 2.5 acres, comprising 33 land-use classes. Because the Denver HOLC map was created in 1938, we can understand the pre-existing patterns of land use before any possible influence by HOLC and identify how those patterns matched how the HOLC categories were appraised from each HOLC area’s description sheet.

We first determined the area and percentage of land-use classes found in each HOLC category in 1937, 1957, 1977, and 1997. For visual aid to the reader, we have changed the original colors used by HOLC (A = Green, B = Blue, C = Yellow, D = Red), to a color-blind friendly palette, while keeping the D category red (A = Dark Blue, B = Light Blue, C = Orange, D = Red)). We overlaid HOLC maps on the land-use data for each decade and used the Tabulate Area tool in ArcMap 10.8 (Esri Redlands) to determine the proportion of each land-use class within a HOLC area. Both HOLC and land-use data were processed in projected coordinate system NAD 1983 UTM Zone 13N. We examined land-use patterns as the diversity of land use in each HOLC category over time, using metrics of alpha diversity (the Shannon-Wiener index [43]). Although a foundational metric in information theory, the Shannon-Wiener index is also often used in ecological literature to describe biological diversity, and has been applied to studies examining the landscape patterns of urban land use and land cover, both at the city [44,45] and neighborhood scales [46]. In our application, the Shannon-Wiener index (H’) described both the proportion of land-use class within an area, and the relative frequency of those land-use classes.

A high H’ value indicates more land-use classes evenly distributed within a HOLC category, while a low value indicates relatively few land-use classes, which may be distributed in uneven groupings. H’ is resistant to configurational shifts within an area, as compared to other urban land-use diversity metrics [47], which makes H’ more suitable for analyzing the overall change in land-use composition at a neighborhood scale. We tested for statistical significance in the HOLC categories and Year of Analysis influence on Shannon Index through a two-way Analysis of Variance test (ANOVA), and we tested statistical significance between specific HOLC categories using Tukey’s Honest Significant Differences (HSD) test, a method regularly used with analyzing post-hoc comparisons among HOLC categories [17,48]. We used a threshold of p < 0.05 throughout to determine significance.

To extend the time-series into the current era (~2020), we first added to the Drummond et al. (2019) land-use dataset [9] with a recent land-use dataset from 2018 that used the same resolution (1 m) to allow additional change analyses between 1997 and 2018, a period of rapid urban growth[49]. However, because Drummond et al. [9] and Chen et al. [49] used different classification methodologies and land-use naming conventions, we only included the comparison of residential land use. Secondly, we analyzed the potential influence of zoning, both on current land-use configuration and how zoning changes are distributed across in formally HOLC-designated areas. Zoning data were provided by the City of Denver, and although these data do not directly sync up to our demographic and land-use time-series, they do provide current patterns of zoning designation, as well as the dates at which zoning changes occurred within a parcel dating from 1955–2019. However, the zoning data do not provide a description of what the designation was changed from.

Analysis of demography time series

We examined demography related to HOLC redlining by deriving data on race, education, and income from the 1940, 1960, 1980, and 2000 U.S. Census through IPUMS [10]. We used these years as the decennial Census events most closely matching years for which we had land-use data. To examine the most current state of demographic patterns in Denver, we used the 2015–2019 American Community Survey [50]. We used weighted tract-level percentage of non-White population as our key racial demographic descriptor, as racial descriptions that specifically identify Black, Indigenous, and other People of Color have changed over the course of the decennial Census and are difficult to compare across decades. For education, we used the tract-level percentage of college-educated individuals over 25 years old. To extract HOLC-level data, we overlaid all tract-level demographics by the HOLC maps and spatially joined to the Census tract level for each decennial Census. Then, using the Tabulate Intersection tool in ArcMap 10.8, we extracted the area of each Census tract contained by a HOLC area. We used this area as a weighting value to calculate the demographic percentages within each HOLC area.

Because the 1940 decennial Census did not include tract-level data on number of households within income brackets, we limited the income analyses to the 1960, 1980, 2000 decennial Census and 2015–2019 American Community Survey. This approach allowed for the comparison of the number of households within income brackets across 59 years. We classified household income into “low,” “medium,” and “high” categories based on The Pew Research Center classification of “Middle-Income” being 67%-200% of the median household income for the region and year, which also precluded the need to adjust for inflation across years [51]. We determined significant differences for education, race, and income class across HOLC categories and years using a generalized linear model (glm) with interactions between HOLC category and years (and income class for income demographics). Models were fit with a quasibinomial distribution to account for the potential clustering of data within HOLC categories, with independent variables of HOLC category and year, and the demographic proportion being our dependent variable. Because education, race, and income are known to be correlated with each other, we modeled each demographic factor independent from each other. This resulted in five unique models: HOLC category and Year effects on non-White residents; college graduates; and low, middle, high income households. We determined significance of interactions between demographic data and HOLC category through p-values with an alpha of 0.05.

Analysis of gentrification

We examined connections to land use and patterns of gentrification by using a previously developed “gentrification index” that determines whether a Census tract has gentrified or not between the years of 2000 and 2019 [52,53]. A census tract is determined to be eligible for gentrification if its median income and home value was within the bottom 40^th percentile compared to tracts within the same metro area during the year 2000. This eligibility identifies groups of lower-income neighborhoods with below average household values. A neighborhood was then determined to be “gentrified” if it met all the following criteria: [1] the median home value had increased when adjusted for inflation within the study period (in this case 2000–2019), and [2] the inflation-adjusted median home value placed the tract in the top one-third of all tracts within the metro area in the year 2019, and [3] the increase in college-educated adults (holding a minimum of a bachelor’s degree) was also in the top one-third of tracts within the same metro area in the year 2019. Using the Tabulate Intersection tool in ArcMap 10.8, we determined the percentage of each HOLC area covered by gentrified tracts, tracts eligible for gentrification, and tracts ineligible for gentrification. To determine demographic influence on total gentrified area for each HOLC category, we conducted linear regressions using the percentage of non-White population as an independent variable. Because the gentrification index itself is calculated using education and income data, we did not use those demographic inputs as independent variables in our analysis of gentrification drivers. We completed this analysis using the original methodology of the “Gentrification Index” from Freeman [52], the Longitudinal Tract Database developed by Logan et al. [54], and Python code developed by Vo [55]. All statistical analysis was completed in R Version 4.2.2 [56].

Results

Land-use diversity

We found HOLC D categories exhibited a greater diversity of land-use classes in 1937, and it continued to have higher diversity of land-use classes over time compared to HOLC categories A and B (Fig. 2A). Although land use in HOLC A categories was primarily used for single-family residences in 1937, by 1997 single-family land use in HOLC A categories had both increased by the largest percentage (25%) and was the most dominant land-use class (93%) compared to other HOLC categories (Fig. 2A,Table 1).

Fig. 2 — A. Visual description of land-use classes of HOLC categories at 20-year periods between 1937 and 1997. Percentage values are listed in Table 1. B: Boxplots comparing land-use Shannon Diversity of each HOLC category at each time-period. Significant differences between groups are represented by different letters above boxplots. The boxes represent the interquartile range (IQR) of values, the horizontal line within each box indicates the median. Outliers are shown as individual points outside the whiskers, which extend to the value nor further than 1.5 times the IQR from the box. C: Differences in Shannon Diversity visualized.

Table 1. Land use percentages aggregated to Home Owners’ Loan Corporation (HOLC) categories from 1937-1997 in Denver Colorado.

	A				B				C				D
Land Use Type	1937	1957	1977	1997	1937	1957	1977	1997	1937	1957	1977	1997	1937	1957	1977	1997
Agricultural business, general	0%	0%	0%	0%	2%	2%	0%	0%	2%	4%	0%	0%	5%	5%	0%	0%
Bare, transitional	0%	0%	0%	0%	0%	5%	0%	11%	3%	3%	0%	0%	1%	3%	3%	3%
Communication and Utilities	0%	0%	0%	0%	0%	2%	1%	1%	0%	0%	0%	0%	0%	1%	4%	4%
Cropland, irrigated	2%	0%	0%	0%	0%	0%	0%	0%	7%	0%	0%	0%	11%	6%	0%	0%
Cropland, non-irrigated	7%	0%	0%	0%	0%	0%	0%	0%	5%	0%	0%	0%	4%	3%	3%	3%
Entertainment and recreation	4%	0%	0%	0%	2%	0%	0%	0%	2%	3%	3%	3%	2%	2%	4%	3%
Forest	0%	0%	0%	0%	1%	0%	0%	0%	0%	2%	0%	0%	2%	0%	0%	0%
Golf Course	0%	0%	0%	0%	5%	5%	4%	4%	0%	0%	0%	0%	0%	0%	0%	2%
Grassland/ Herbaceous	0%	2%	0%	0%	0%	2%	4%	0%	0%	5%	0%	0%	0%	10%	4%	3%
Industrial, heavy	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	3%	3%	3%	3%
Industrial, light	0%	0%	0%	0%	0%	0%	0%	0%	3%	0%	0%	0%	22%	15%	23%	22%
Institutional, general	0%	1%	1%	1%	3%	2%	3%	3%	3%	3%	5%	5%	2%	3%	3%	3%
Mixed Urban	5%	0%	0%	0%	5%	6%	1%	1%	15%	17%	17%	18%	14%	20%	14%	14%
Office	0%	0%	0%	0%	3%	3%	3%	0%	10%	6%	5%	4%	0%	2%	4%	4%
Orchards and groves	0%	0%	0%	0%	0%	0%	0%	0%	3%	0%	0%	0%	0%	0%	0%	0%
Quarries/ Gravel Pits	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	6%	4%	0%	0%
Reservoirs	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	3%	2%	0%	0%
Residential, multifamily	0%	3%	3%	3%	3%	8%	6%	6%	3%	7%	8%	9%	3%	5%	8%	9%
Residential, single family	68%	89%	93%	93%	78%	85%	85%	85%	67%	74%	71%	70%	49%	51%	46%	45%
Retail and services, major	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	2%	2%	0%	5%	1%	5%
Retail and services, minor	0%	1%	2%	2%	5%	6%	7%	7%	7%	8%	10%	10%	7%	8%	11%	12%
School	2%	3%	3%	3%	2%	3%	3%	3%	2%	3%	3%	3%	3%	3%	5%	5%
Stream and river	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	0%	2%	2%	2%	2%
Transportation, ground	12%	12%	7%	7%	3%	3%	4%	4%	3%	3%	3%	3%	5%	5%	8%	8%
Urban Park	0%	3%	3%	3%	0%	2%	2%	2%	0%	2%	2%	2%	0%	3%	4%	4%

Open in a new tab

At the other end of the spectrum, HOLC D categories not only exhibited the lowest amount of single-family residential land use in 1937 (49%), but that value decreased to 45% by 1997. In addition, in 1937 HOLC A, C, and D categories all had agricultural land use (A: 9%, C: 12%, D: 20%) (Table 1). However, by 1957 agricultural land use was not found in use within HOLC A, with a corresponding increase in the area covered with single-family homes. In HOLC D, agricultural land did not drop below 10% until 1977, by which time single-family residential area had decreased, while small increases occurred in multiple commercial and industrial land uses. When comparing residential land-use data from 2018 (40%) to that from 1997 (19%), we found an increase in high-income households was positively correlated with an increase in residential land use in only HOLC D neighborhoods (S1 Fig).

Overall, land-use diversity within each HOLC category followed distinct patterns. For all decades of land-use diversity analysis we found the HOLC category had a significant effect on Shannon Diversity (1937: F = 9.44, DF = 3, p-value < 0.0001; 1957: F = 13.67, DF = 3, p-value < 0.0001; 1977: F = 18.53, DF = 3, p-value < 0.0001; 1997: F = 16.23, DF = 3, p-value < 0.0001). In 1937 only HOLC D was significantly different from all other HOLC categories, but by 1977, both HOLC categories C and D were significantly distinct from each other, and from HOLC A (Fig. 2B). HOLC D categories exhibited an increase in land-use alpha diversity every 20 years, while HOLC A displayed a decreasing pattern in land-use diversity over the same period (Fig. 2C). When analyzing zoning data, we found areas formally designated HOLC-D had significant higher percentages of areas zoned for commercial, industrial, and mixed-use compared to former HOLC A areas (S2A Fig). Furthermore, we found the HOLC C and D categories have a significantly greater total of zoning changes from 1955 to 2019, and that these changes have occurred incrementally during that entire period (S2B Fig and 2C).

Demographics

From 1940 onward, patterns of race and education across HOLC categories were similar to patterns of land-use diversity (Fig. 3A and B). The percentage of non-Whites both increased with year and HOLC Category, with an interaction effect between those two factors (HOLC category: F = 20.84, DF = 3, p-value < 0.001; Year: F = 34.31, DF = 4, p-value < 0.001; HOLC * Year: F = 1.92, DF = 12, p-value = 0.03). The percentage of non-Whites increased from HOLC A to HOLC D categories (Table 2), increased overall in Denver from 1940 to 2019, and was significantly higher in HOLC D than HOLC A and B from 1980 to 2019. The proportion of college-educated individuals increased as a factor of year, of HOLC category, and an interaction effect between the two factors (HOLC category: F = 72.09, DF = 3, p-value < 0.001; Year: F = 155.09, DF = 4, p-value < 0.001; HOLC * Year: F = 3.17, DF = 12, p-value < 0.001). This means that while the college-educated populations decreased from HOLC A to D categories and the overall number of college-educated populations also increased from 1940 to 2019, the size of that increase was significantly smaller in HOLC D areas in 1980, 2000, but significantly larger in 2019, compared to other HOLC categories. Within decades, HOLC categories did not differ statistically in the proportions of non-White population in 1940 or 1960, although by 1980, HOLC D categories had statistically significant higher non-White proportions. HOLC D also had significantly lower college-educated populations compared to HOLC A for the entire study period.

Table 2. Percent changes to demographic metrics in Home Owners’ Loan Corporation (HOLC) categories. Education and racial data start in 1940 and income data start in 1960.

Demographic Metrics	A				B				C				D
Demographic Metrics	1940–1960	1960–1980	1980–2000	2000–2019	1940–1960	1960–1980	1980–2000	2000–2019	1940–1960	1960–1980	1980–2000	2000–2019	1940–1960	1960–1980	1980–2000	2000–2019
College Educated Fraction	29%	90%	33%	15%	47%	94%	69%	48%	35%	111%	84%	78%	56%	245%	102%	147%
Non-White Fraction	2%	1463%	71%	22%	201%	2408%	48%	-5%	1279%	2016%	42%	-21%	1091%	3626%	22%	-41%
High-Income Households	na	-43%	86%	5%	na	-62%	220%	49%	na	-69%	296%	87%	na	-76%	612%	58%
Medium-Income Households	na	14%	-26%	2%	na	-5%	-15%	-7%	na	-36%	17%	0%	na	-34%	28%	7%
Low-Income Households	na	109%	-36%	-5%	na	172%	-36%	-20%	na	147%	-30%	-23%	na	156%	3%	-18%

Open in a new tab

The proportion of households in different income classes varied across year and HOLC category (Fig. 4). In low- and high-income households, we observed a significant effect of HOLC category and year on proportion of low-income households, but no significant interaction between HOLC category and year (Low Income - HOLC category: F = 59.87, DF = 3, p-value < 0.001; Year: F = 94.45, DF = 3, p-value < 0.001; HOLC * Year: F = 1.66, DF = 9, p-value = 0.10. High Income - HOLC category: F = 66.42, DF = 3, p-value < 0.001; Year: F = 48.16, DF = 3, p-value < 0.001; HOLC * Year: F = 1.28, DF = 9, p-value = 0.252). For middle-income households, we observed a significant interaction between HOLC category and year (HOLC category: F = 3.55, DF = 3, p-value = 0.016; Year: F = 45.12, DF = 3, p-value < 0.001; HOLC * Year: F = 5.64, DF = 9, p-value < 0.0001). During 1980, the HOLC C and D category increased in the proportion low-income households significantly more than HOLC A or B (HOLC C – Year 1980: p-value = 0.005; HOLC D – Year 1980: p-value = 0.001). Within decades, in 1960, high- and middle-income households displayed differences between HOLC categories, with HOLC C and D categories having significantly lower proportions of middle and high-income households, and middle-income households. Although there was no difference between low-income household proportion in 1960, in 1980 and the decades following, we found significantly higher proportions of low-income households in HOLC D categories (Fig. 4).

The directions and magnitude of proportional changes in our demographic metrics varied depending on year and variable. While every 20-year period displayed an increase in the college-educated proportion of population, the magnitude of change was generally always largest in HOLC D, decreasing in magnitude to HOLC A, with the largest magnitude of change in the latest period. The non-White proportion of population also increased across all HOLC categories up to the year 2000, with very large increases from 1940 to 1980, where there was a proportional decline in non-White residents in HOLC B, C, and D categories, with HOLC D experiencing the largest proportional decline. Income metric patterns also varied by variable. The proportional change to high-income households maintained the same direction of change across all HOLC categories for each period, with increases in high-income households occurring steadily after 1980 (Fig. 4 and Table 2). For the medium- and low-income variables, the patterns varied but magnitude differences across HOLC categories remained relatively low. These patterns indicate that from 1940 onward, Denver has attracted college-educated and racially diverse residents across the city, with patterns in income following similar macro-citywide patterns. However, in the last 20 years, the HOLC C and D categories have seen the largest decrease in diverse populations, as well as the largest increases in college-educated residents and high-income households.

Gentrification

According to our “gentrification index,” between 2000 and 2019, 100% of HOLC A areas were ineligible for gentrification, with decreasing areas of ineligibility occurring in HOLC’s B through D. HOLC D contained significantly more area eligible for gentrification in 2019 compared to all other HOLC categories, although no difference was found between HOLC categories and percent area gentrified (Fig. 5A). When examining racial characteristics as a driver of gentrification within neighborhoods eligible for gentrification, higher proportions of non-White population predicted higher percentages of gentrification only in HOLC C categories (p-value = 0.001, Fig. 5B).

Fig. 5 — *A: The 2019 area percentage in HOLC category that is either ineligible for gentrification (household income and median home value were above the 40*^th *percentile within the metro area in 2000), eligible for gentrification (household income and median home value were in the bottom 40*^th percentile within the metro area in 2000), or gentrified (eligible tracts from 2000 experienced an increase in home value by 2019, and median income and percent college-educated residents were within the top one-third for the metro area). The middle point corresponds with the data mean and whiskers corresponds to the standard error for data within the HOLC category and gentrification variable. A significant difference between groups is represented by different letters. B. The relationship between the percentage of non-White residents in the year 2000 and the percent area gentrified in all HOLC areas eligible for gentrification in 2019. Shaded area represents the estimated standard error. We derived data from the U.S. decennial Census through IPUMS, the 2019 American Community Survey and the Longitudinal Tract Database [10,50,54].

Discussion

Our aim was to address the gaps in knowledge regarding the legacy of social and land-use patterns varied across HOLC categories over the past century. By pairing unique high-resolution land-use data with demographics from the U.S. Census, we show how historical redlining not only illustrated differing patterns of land use and demography prior to the creation of Denver’s HOLC map, but also how these patterns continued to evolve over time. Certain differences between HOLC A and HOLC D areas in Denver became statistically significant both in the years prior to redlining and in the decades following, as we observed significant differences of land-use diversity in 1937, but not until 1980 (after the practice of redlining was made illegal by the Fair Housing Act of 1968) for demographics of race and low-income status. Also, the evidence of significant segregation in land-use composition, education, and high-income residents dating to the creation of Denver’s HOLC map adds weight to the argument that segregation-based lending practices incorporated pre-existing segregation and reinforced it after these practices were made illegal [7]. These results illustrate the multi-decadal legacy of urban segregation and the potential reinforcement of that segregation through discriminatory housing policies and neighborhood reinvestment. However, between 2000 and 2019, shifts in some of the demographic patterns emerged in demographics and land use, indicating how long-term patterns can be rapidly changed by gentrification.

Drivers of land-use diversity

Various regulatory mechanisms—such as zoning, restrictive covenants, barriers to funding, and special districts—influence land-use changes [57–59]. Unraveling the cause and effect of these mechanism is beyond the scope of this paper; however, the HOLC boundaries provide a fixed spatial extent in which descriptive information about racial and ethnic diversity and economic indicators were standardized nationally and shed light on the pre-existing conditions of a city prior to mortgage lending practices having an impact [7,13]. Our data corroborates this: prior to 1940, an inequality of land-use diversity, higher-income households, and college-educated populations already existed across Denver’s HOLC areas. Our evidence supports Hillier’s 2003 hypothesis that HOLC practices are not causes of present-day inequality, but reinforced segregation through “racial steering” [13]. Further comparing HOLC areas over time, we found significant differences in the Shannon-Wiener index as early as 1937 followed by differing patterns of land-use diversity across HOLC groups in subsequent decades. HOLC A areas declined in alpha diversity after 1937, and then stabilized after 1977. However, in HOLC C and D, the land-use alpha diversity continued to increase post-1937. This change in land-use diversity is reflected in the significant amount of multiple zoning categories in formally HOLC D areas compared to formally HOLC A (S2A Fig), and the regular changes in zoning occurring in HOLC C and D between 1960 and 2020 (S2B Fig and 2C). Detailed information about directional changes in zoning ordinances (specifically knowing what the zoning was before a change), restrictive covenants, or special districts that may have affected these patterns were excluded from our analysis because they could not provide the same level of standardized descriptive information or fixed spatial extents.

Given that the drivers of land-use diversity are multi-faceted and complex, we focus on understanding the balance between the fixed or flexible nature of land use in HOLC areas over time. Although we do not have records of where FHA and HOLC loans were distributed in Denver in the 1930s, the fixed versus variable presence of single-family homes across HOLC categories indicates the lasting impact of denying low-cost mortgage insurance in HOLC D categories. Single-family residential zoning is the most fixed of any land-use type, once a single-family residential area is established, it almost always remains as single-family zoning [60,61]. For areas with detached homes, an increase in homeownership rates can create the demand for the area to be converted to single-family zoning [62,63]. Because potential homeowners HOLC D were prevented from achieving homeownership, these neighborhoods did not have the chance to establish single-family zoning and would have a higher chance of land use diversifying over time. Our results then provide evidence for the permanence of single-family residential areas: by 1957 HOLC classes A-C had increases in single family residential compared to HOLC D. Moreover, post 1957, residential land uses in HOLC A and B remained stable, whereas in HOLC D after 1957 there is a small decline of the percentage of single-family homes and a steady increase in multi-family homes and other non-residential land-use classes (Table 1). The fluctuation in single-family homes and greater overall land-use diversity within redlined districts (Fig. 2) implies HOLC D categories were more vulnerable to changes in land-use regulation. Thus, the land-use diversity in HOLC D categories appears to be the result of a combination of micro- and macro-scale influences.

The city-wide urban renewal polices during the 1960s through 1980s resulted in substantial land-use and demographic changes in downtown districts. Although the downtown districts themselves—which had relatively minimal residential land use when the Denver HOLC maps were made—were not assigned categories, the downtown is surrounded by formally HOLC D categories. DURA was created in 1958 with the aim of attracting more business to Denver’s downtown and creating more public amenities within generally low-income, Hispanic neighborhoods [39]. Two of DURA’s largest projects were the Skyline and Auraria projects, which involved the removal of “blighted” housing and relocation of mostly lower-income Hispanic residents, and both projects faced strong public resistance before their eventual adoption [64,65]. Although these urban renewal efforts, along with subsequent smaller downtown land-use developments during the 1980s through early 2000s (e.g., Coors Field, The Denver Performing Arts Center, Confluence Park, and The Convention Center), did not directly occupy redlined areas, the rapid redevelopment and increase of amenities within downtown attracted young educated individuals to the affordable locations in adjacent HOLC D districts [65]. Consequently, all three HOLC D areas that have gentrified surround these downtown urban renewal developments (HOLC D9, D10, and D12).

Shifts in land-use patterns and demographics

The historical legacy of diversity of land-use composition within initially described “declining” areas of Denver may result in the development of new residential land uses to attract higher-income earners, at the possible expense of often lower-income long-term residents. Our data alone, however, does not state whether displacement is occurring or has occurred. Rather, we show how the built environment has been fixed in some areas (HOLC A/B) and flexible in others (HOLC C/D). That flexibility in built environment is seen in the loss of single-family homes in formally HOLC D neighborhoods between 1957 to 1997 and the increases in multifamily homes, light industrial, and minor retail services during the same period. Those changes in land use are mirrored by dynamic shifts in demographics, such as the proportional loss in high- and middle-income households in formerly HOLC D areas between 1960 and 1980. Then in the following two decades, formerly HOLC D areas had the largest percentage increase in high-income households and college-educated residents (Table 2). More research would be useful understand these shifts in income and demographics. Given the staying power of single-family residential zoning, it is possible that HOLC A/B categories simply cannot densify at the same rate as HOLC C/D categories due to the large proportion of single-family residential zoning. However, given the vulnerabilities of low-income and non-White communities to displacement, understanding whether these changes reflect an increase in residents overall or an exodus of low-income and non-White households is an important area for future research.

Understanding the “chicken or egg” scenario of whether urban redevelopment opportunities drive demographic shifts, or if land-use changes are taken advantage of after demographic changes occur, is an active area of study [25,26]. Moreover, studies have opined that segregation-based housing policies may have directly influenced the rise of gentrification in the last 20 years [39,66]. In Denver, we found that the land-use patterns within HOLC A categories were stable after 1960, but HOLC C and D areas had greater diversity of land use during all periods, exhibiting a larger variation of residential and urban developed classes. A similar outcome was found in post-apartheid South Africa, where the Brixton neighborhood of Johannesburg has become more ethnically diverse in the years following the end of apartheid (1996–2011), and has shown evidence of becoming a desirable place for professional investment, but at the time of analysis (2017) did not show evidence of gentrification [20].

Land-use and demographic drivers of gentrification

The gentrification that results from established patterns of land use and demographics may take many years to become noticeable. For example, we observed a general increase in proportion of White and college-educated populations in HOLC C and D areas in 2019 (Fig. 3), and this increase in these demographics to areas HOLC initially described as “declining” or “hazardous” acts as a potential indicator of gentrification. Although we found a statistical relationship between race and gentrification in formerly HOLC C but not HOLC D areas, the fact that the formerly HOLC D category has significantly more “gentrification-eligible” area could indicate if these demographic patterns continue, more formerly HOLC D areas may become gentrified. We expect this is likely to occur because the recent finding that neighborhoods with higher incomes (like the formerly HOLC C compared to HOLC D areas) are more likely to gentrify [28]. As indicated in our results, the proportion of high-income homes has grown in formerly HOLC D areas (Fig. 4). This result, combined with the high number of gentrification-eligible areas, indicates that continued gentrification of some formerly HOLC D areas is likely.

Our gentrification analyses were somewhat limited because HOLC areas that may have gentrified before 2000 would show up as ineligible for gentrification. For example, HOLC D2 (the Cherry Creek neighborhood) was described in the HOLC documentation as “Developers and real estate men have looked forward to buying cheap houses in this area for newer development. [1]” New investment did come to Cherry Creek after the construction of the Cherry Creek Dam and the subsequent Cherry Creek Shopping Center in the early 1950s, followed by the center’s $250 million revitalization in 1990 [67]. Although gentrification can occur due to redevelopment of land use and increased construction of residential housing for wealthier newcomers to a neighborhood, more affluent White individuals may also move into pre-existing, lower-cost homes in HOLC C and D categories, replacing previous residents who are lower-income and People of Color [68]. This top-down land-use development and bottom-up demographic displacement are likely occurring in tandem and reinforcing each other. We find suggestive evidence of both phenomena, for example a bottom-up demographic replacement being where a larger proportion of non-White residents indicated a greater percentage of formerly HOLC C areas to be gentrified (Fig. 5), and a potential top-down linkage between land use and demographics being the increase in higher-income households in formerly HOLC D areas correlating with an increase in residential land use (S1 Fig).

In the United States, although HOLC maps described pre-existing land-use and demographic segregation, we found that further patterns of segregation continued to develop over time but can also quickly change as observed between 2000 and 2019 during periods of rapid growth and gentrification in Denver. The patterns we observed are likely not unique to Denver. Redlining policies were put in place in many cities across the United States. Patterns of gentrification over the past 20 years have also been observed in multiple cities with a history of redlining [69]. However, with the multiple studies examining the present-day effects of redlining practices, current sociological, economic, public health, and ecological patterns resulting from redlining are not the same in all cities [70]. For example, recent work has found that in the city of Portland, Oregon, urban tree canopy cover actually decreased in former HOLC A while increasing in HOLC B – D; while over the same period in Philadelphia, Pennsylvania, tree canopy decreased over all HOLC categories [22]. Moving beyond the United States, while other nations exhibit historical and present institutionalized exclusionary housing policies [20,34], un-official discriminatory housing actions are found in cities worldwide, and can play a part in establishing and re-enforcing land-use and demographics patterns of discrimination [35,71].

Limitations

The purpose of our study and research question was to resolve uncertainty around HOLC’s influence on land use and demographic changes over the past 80 years and examine if those patterns help explain where gentrification occurred in the city of Denver. This is a novel contribution to research on redlining because it used a unique time-series dataset of high-resolution land use for the city of Denver. This highlights an important limitation: land-use findings cannot be replicated for other cities without the presence of such a dataset. Data availability is often a limitation for redlining studies aiming to track change through time, and such studies often require a substantial amount of work just in recovering historical data. Or, in the case of including multiple drivers to land-use change, finding and digitizing additional land-use regulation data such as zoning, racial covenants, or special districts that may be available for only certain time periods is challenging. On the other hand, our use of demographics highlights a publicly available source of time-series data (IPUMS National Historic Geographic Information System at the University of Minnesota (www.nhgis.org)) that this study has shown is significantly tied to pre-HOLC segregation and to the legacies that followed over the decades [10]. Thus, this research presents a possible path forward for similar studies in other cities.

Conclusion

We came to the study with the goal of finding patterns in the multidecadal data of land use and demographics that could help explain present day patterns of gentrification in a U.S. city. By using the HOLC boundaries, we were able to track how the segregation in land-use diversity, housing, and demographics was established prior to redlining practices in Denver, but continued to grow in the following 80 years. Redlining was a key aspect of this divergence in patterns, but urban renewal policies and projects of the mid to late 21^st century also reinforced land-use and demographic shifts. These projects, along with urban redevelopment brought more educated and wealthy individuals to formally HOLC C and D areas, which may have been able to accept these demographic changes due to diverse and changing land use patterns. Although our study does not address the issues of demographic replacement caused by gentrification, the patterns we observe do answer questions about how present-day patterns in urban land cover and land use can both develop over many decades, but also be primed to have large shifts towards gentrification if certain conditions are met. Moreover, although the discriminatory housing policies of the 21^st century are not unique to Denver, Colorado, or even the United States, our study was able to highlight uniquely city-specific drivers of change from the past decades. Future studies examining redlining’s impacts could consider how discriminatory patterns develop over many decades as well as examining what the present-day affects may be.

Disclaimer

Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Supporting information

S1 Fig. Comparison of increases in residential land use within Home Owners’ Loan Corporation (HOLC) categories between 1997 and 2018, and increases in high-income households within the same neighborhoods over the same period in Denver, Colorado.

The 1997 and 2018 land-use data are from two different sources (1997: Drummond et al. 2019; 2018: Chen et al 2022) with differing classification methodologies. Only significant relationships have statistics and regression lines displayed. Shaded area represents the estimated standard error.

(TIF)

pone.0317988.s001.tif^{(971.9KB, tif)}

S2 Fig. The influence of zoning ordinances, and their changes on land use configuration across former Home Owners’ Loan Corporation (HOLC) categories in Denver, Colorado.

A: A comparison of the Zoning Class Percentages for six difference zoning designations across HOLC categories in the year 2020. B: The total area of parcels in the year 2020, which experienced a zoning change from 1955 to 2020. C: The total sum of area which experienced a zoning change across HOLC categories, aggregated to the decadal scale. The middle point corresponds with the data mean and whiskers corresponds to the standard error for data within the HOLC category and zoning variable. Significant differences between groups are represented by different letters above boxplots.

(TIF)

pone.0317988.s002.tif^{(1MB, tif)}

Data Availability

All relevant data are within the paper and its Supporting Information files and are also available on the website https://www.sciencebase.gov/catalog/item/6712935ed34eb6a152fc6d2a or doi:10.5066/P1RJMRQ5

Funding Statement

The author(s) received no specific funding for this work.

References

1.Nelson RK, Winling L. Mapping Inequality: Redlining in New Deal America [Internet]. 2023 [cited 2022 Dec 12]. Available from: https://dsl.richmond.edu/panorama/redlining
2.Huang SJ, Sehgal NJ. Association of historic redlining and present-day health in Baltimore. Liu SY, editor. PLoS ONE. 2022. Jan 19;17(1):e0261028. doi: 10.1371/journal.pone.0261028 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Locke DH, Hall B, Grove JM, Pickett STA, Ogden LA, Aoki C, et al. Residential housing segregation and urban tree canopy in 37 US cities. npj Urban Sustainability. 2021;1(1):15. [Google Scholar]
4.Nardone AL, Casey JA, Rudolph KE, Karasek D, Mujahid M, Morello-Frosch R, et al. Associations between historical redlining and birth outcomes from 2006 through 2015 in California. Khan HTA, editor. PLoS ONE. 2020. Aug 7;15(8):e0237241. doi: 10.1371/journal.pone.0237241 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Lane H, Morello-Frosch R, Marshall J, Apte J. Historical redlining is associated with present-day air pollution disparities in U.S. cities. Environ Sci Technol Lett. 2022;9(4):345–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Moxley D, Fischer B. Historic HOLC redlining in Indianapolis and the legacy of environmental impacts. Journal of Public and Environmental Affairs [Internet]. 2020. [cited 2024 Jan 29];1(1). Available from: https://scholarworks.iu.edu/journals/index.php/jpea/article/view/30321 [Google Scholar]
7.Fishback P, Rose J, Snowden KA, Storrs T. New evidence on redlining by federal housing programs in the 1930s. J Urban Eco. 2022;103:462. [Google Scholar]
8.Markley S. Federal ‘redlining’ maps: a critical reappraisal. Urban Stud. 2023. doi: 10.1177/00420980231182336 [DOI] [Google Scholar]
9.Drummond MA, Stier MP, Diffendorfer J (Jay) E. Historical land use and land cover for assessing the northern colorado front range urban landscape. J Maps. 2019;15(2):89–93. doi: 10.1080/17445647.2018.1548383 [DOI] [Google Scholar]
10.Ruggles S, Flood S, Sobek M, Backman D, Chen A, Coope G, et al. IPUMS USA: Version 15.0 [Internet]. Minneapolis, MN; 2024. Available from: doi: 10.18128/D010.V15.0 [DOI] [Google Scholar]
11.Mauck LM. Five points neighborhood of Denver. Arcadia Publishing; 2001. [Google Scholar]
12.Verell BD. Spring valley, Washington DC: Changing land use and demographics from 1900-2000. Geographical Bulletin - Gamma Theta Upsilon. 2008;49(2):103–19. [Google Scholar]
13.Hillier AE. Redlining and the home owners’ loan corporation. J Urban Hist. 2003;29(4):394–420. doi: 10.1177/0096144203029004002 [DOI] [Google Scholar]
14.Hyra DS. Conceptualizing the new urban renewal. Urban Aff Review. 2012;48(4):498–527. doi: 10.1177/1078087411434905 [DOI] [Google Scholar]
15.Teaford JC. Urban renewal and its aftermath. Housing Policy Debate. 2000;11(2):443–65. [Google Scholar]
16.Tiagi R. Occupational mobility of black migrants in the West during the 1950s. Labor His. 2015;56(1):58–68. doi: 10.1080/0023656x.2015.991563 [DOI] [Google Scholar]
17.Grove M, Ogden L, Pickett S, Boone C, Buckley G, Locke DH, et al. the legacy effect: understanding how segregation and environmental injustice unfold over time in baltimore. Ann Am As Geogr. 2017;108(2):524–37. doi: 10.1080/24694452.2017.1365585 [DOI] [Google Scholar]
18.Horr A, Hunkler C, Kroneberg C. Ethnic discrimination in the german housing market. Zeitschrift für Soziologie. 2018;47(2):134–46. doi: 10.1515/zfsoz-2018-1009 [DOI] [Google Scholar]
19.Aalbers MB. Place-based and race-based exclusion from mortgage loans: evidence from three cities in the Netherlands. J Urban Aff. 2007;29(1):1–29. doi: 10.1111/j.1467-9906.2007.00320.x [DOI] [Google Scholar]
20.Haferburg C, Huchzermeyer M. Redlining or renewal? The space-based construction of decay and its contestation through local agency in Brixton, Johannesburg. In: Kirkness P, Tije-Dra A, editors. Negative Neighbourhood Reputation and Place Attachment: The Production and Contestation of Territorial Stigma [Internet]. 1st ed. Routledge; 2017. [cited 2023 Sep 24]. Available from: https://www.taylorfrancis.com/books/9781317089537 [Google Scholar]
21.Wilson B. Urban Heat Management and the Legacy of Redlining. J Am Plan As. 2020;86(4):443–57. doi: 10.1080/01944363.2020.1759127 [DOI] [Google Scholar]
22.Jung MC, Yost MG, Dannenberg AL, Dyson K, Alberti M. Legacies of redlining lead to unequal cooling effects of urban tree canopy. Landscape Urban Plan. 2024;246:105028. doi: 10.1016/j.landurbplan.2024.105028 [DOI] [Google Scholar]
23.Walker RH, Keeler BL, Derickson KD. The impacts of racially discriminatory housing policies on the distribution of intra-urban heat and tree canopy: a comparison of racial covenants and redlining in Minneapolis, MN. Landscape Urban Plan. 2024;245:105019. doi: 10.1016/j.landurbplan.2024.105019 [DOI] [Google Scholar]
24.Donovan GH, Prestemon JP, Butry DT, Kaminski AR, Monleon VJ. The politics of urban trees: tree planting is associated with gentrification in Portland, Oregon. For Policy Econ. 2021;124:10.1016/j.forpol.2020.102387. doi: 10.1016/j.forpol.2020.102387 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Kolko J. The Determinants of Gentrification. SSRN Electronic Journal [Internet]. 2007. Dec. Available from: http://www.ssrn.com/abstract=985714 [Google Scholar]
26.Anguelovski I, Connolly JJT, Cole H, Garcia-Lamarca M, Triguero-Mas M, Baró F, et al. Green gentrification in European and North American cities. Nat Commun. 2022;13(1):3816. doi: 10.1038/s41467-022-31572-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Atkinson R, Wulff M, Reynolds M, Spinney A. Gentrification and displacement: the household impacts of neighbourhood change. Aust Housing Urban Res Inst. 2011. [Google Scholar]
28.Leguizamon S, Christafore D. The influence of land use regulation on the probability that low-income neighbourhoods will gentrify. Urban Stud. 2020;58(5):993–1013. doi: 10.1177/0042098020940163 [DOI] [Google Scholar]
29.Curran W. From the frying pan to the oven: Gentrification and the experience of industrial displacement in Williamsburg, Brooklyn. Urban Stud. 2007;44(8):1427–40. doi: 10.1080/00420980701373438 [DOI] [Google Scholar]
30.Davis J. How do upzonings impact neighborhood demographic change? examining the link between land use policy and gentrification in New York City. Land Use Policy. 2021;103:105347. doi: 10.1016/j.landusepol.2021.105347 [DOI] [Google Scholar]
31.Gould KA, Lewis TL. Green gentrification: Urban sustainability and the struggle for environmental justice. Routledge. 2016;192. [Google Scholar]
32.Pham T-T-H, Apparicio P, Landry S, Lewnard J. Disentangling the effects of urban form and socio-demographic context on street tree cover: a multi-level analysis from Montréal. Landscape Urban Plan. 2017;157:422–33. doi: 10.1016/j.landurbplan.2016.09.001 [DOI] [Google Scholar]
33.Gibbons J. Examining the long-term influence of New Deal era redlining on contemporary gentrification. Urban Stud. 2023;60(14):2816–34. [Google Scholar]
34.Lazin FA. The housing policies for Ethiopian immigrants in Israel: Spatial segregation, economic feasbility and political acceptability. Nationalism Ethn Polit. 1997;3(4):39–68. doi: 10.1080/13537119708428517 [DOI] [Google Scholar]
35.Singh G, Vithayathil T, Pradhan KC. Recasting inequality: residential segregation by caste over time in urban India. Environ Urban. 2019;31(2):615–34. doi: 10.1177/0956247818812330 [DOI] [Google Scholar]
36.U.S. Census Bureau. Southern and Western Regions Experienced Rapid Growth This Decade [Internet]. United States Census; 2020. Available from: https://www.census.gov/newsroom/press-releases/2020/south-west-fastest-growing.html
37.Bunyak D, Simmons TH, Simmons RL. Denver area post-world war II suburbs. Colorado Dep Trans. 2011;288. [Google Scholar]
38.Abbott C. Suburb and city: changing patterns of socioeconomic status in metropolitan Denver since 1940. Soc Sci Hist. 1977;2(1):53. [Google Scholar]
39.Connor D, Gutmann M, Cunningham A, Clement K, Leyk S. How entrenched is the spatial structure of inequality in cities? evidence from the integration of census and housing data for Denver from 1940 to 2016. Ann Am Ass Geogr. 2020;110(4):1022–39. [Google Scholar]
40.Denver Urban Renewal Authority. 50 years of revitalizing Denver [Internet]. 2008. Available from: https://digital.denverlibrary.org/nodes/view/822540
41.Avila E, Rose MH. Race, culture, politics, and urban renewal: an introduction. J Urban Hist. 2009;35(3):335–47. [Google Scholar]
42.Frey WH. America’s Largest Metro Areas [Internet]. Brookings Mountain West; 2022. p. 1–32. Available from: https://digitalscholarship.unlv.edu/brookings_policybriefs_reports/11 [Google Scholar]
43.Shannon CE. A mathematical theory of communication. Bell Syst Techn J. 1948;27(3):53. [Google Scholar]
44.Liu H, Weng Q. Landscape metrics for analysing urbanization-induced land use and land cover changes. Geocarto Int. 2013;28(7):582–93. doi: 10.1080/10106049.2012.752530 [DOI] [Google Scholar]
45.Sapena M, Ruiz LÁ. Analysis of land use/land cover spatio-temporal metrics and population dynamics for urban growth characterization. Comput, Environ Urban Syst. 2019;73:27–39. doi: 10.1016/j.compenvurbsys.2018.08.001 [DOI] [Google Scholar]
46.Alsavada MO, Karimi K. The Spatio-Functional Role of Navigable Urban Canals in the City: Cases From London and Amsterdam. UP [Internet]. 2023 Aug 17. [cited 2024 Sep 24];8(3). Available from: https://www.cogitatiopress.com/urbanplanning/article/view/6740 [Google Scholar]
47.Yoshida T, Tanaka K. Land-use diversity index: a new means of detecting diversity at landscape level. Landscape Ecol Eng. 2005;1(2):201–6. doi: 10.1007/s11355-005-0022-0 [DOI] [Google Scholar]
48.Hoffman J, Shandas V, Pendleton N. The effects of historical housing policies on resident exposure to intra-urban heat: a study of 108 US urban areas. Climate. 2020;8(1):12. [Google Scholar]
49.Chen B, Tu Y, Song Y, Theobald DM, Zhang T, Ren Z, et al. Mapping essential urban land use categories with open big data: Results for five metropolitan areas in the United States of America. ISPRS J Photogramm Remote Sens. 2021;178:203–18. doi: 10.1016/j.isprsjprs.2021.06.010 [DOI] [Google Scholar]
50.U.S. Census Bureau’s American Community Survey Office. American Fact Finder - 2019 American Community Survey. 2019. US Census Bureau. [Google Scholar]
51.Fry R, Taylor P. The rise of residential segregation by income. Washington, DC: Pew Research Center. 2012;(202):26. [Google Scholar]
52.Freeman L. Displacement or succession?: residential mobility in gentrifying neighborhoods. Urban Affair Rev. 2005;40(4):463–91. [Google Scholar]
53.Maciag Michael. Gentrification Report Methodology. Governing: The Future of States and Localities [Internet]. 2015. Jan 23; Available from: https://www.governing.com/archive/gentrification-report-methodology.html [Google Scholar]
54.Logan JR, Stults BJ, Xu Z. Validating population estimates for harmonized census tract data, 2000–2010. Ann Am Ass Geogr. 2016. Sep 2;106(5):1013–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Vo, Thuy L. NYC 311 complaints and demographic analysis — 2010 to 2018 [Internet]. 2018. Available from: https://github.com/BuzzFeedNews/2018-06-nyc-311-complaints-and-gentrification
56.RStudio Team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: RStudio, PBC; 2021. Available from: http://www.rstudio.com/ [Google Scholar]
57.Lens MC. Zoning, land use, and the reproduction of urban inequality. Annu Rev Sociol. 2022;48(1):421–39. doi: 10.1146/annurev-soc-030420-122027 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Baics G, Meisterlin L. Zoning before zoning: land use and density in mid-nineteenth-century New York City. Ann Am Assoc Geogr. 2016;106(5):1152–75. [Google Scholar]
59.Irwin EG, Geoghegan J. Theory, data, methods: developing spatially explicit economic models of land use change. Agricult, Ecosyst Environ. 2001;85(1):7–24. [Google Scholar]
60.Ellickson RC. The zoning straitjacket: the freezing of American neighborhoods of single-family houses. Indiana Law J. 2020;96:395. [Google Scholar]
61.Brooks L, Schuetz J. Does housing growth in Washington, D.C., reflect land use policy changes?. Cityscape. 2023;25(2):203–24. [Google Scholar]
62.Haurin DR, Dietz RD, Weinberg BA. The impact of neighborhood homeownership rates: a review of the theoretical and empirical literature. J Housing Res. 2002;13(2):119–51. [Google Scholar]
63.Been V, Madar J, McDonnell S. Urban land‐use regulation: are homevoters overtaking the growth machine?. J Empir Leg Stud. 2014;11(2):227–65. [Google Scholar]
64.Page B, Ross E. Envisioning the urban past: GIS reconstruction of a lost Denver district. Front Digit Human. 2015;2. [Google Scholar]
65.Page B, Ross E. Legacies of a contested campus: urban renewal, community resistance, and the origins of gentrification in Denver. Urban Geogr. 2016;38(9):1293–328. doi: 10.1080/02723638.2016.1228420 [DOI] [Google Scholar]
66.Engels B. Capital flows, redlining and gentrification: the pattern of mortgage lending and social change in glebe, Sydney, 1960–1984*. Int J Urban Regional Res. 1994;18(4):628–57. doi: 10.1111/j.1468-2427.1994.tb00290.x [DOI] [Google Scholar]
67.Simmons T, Simmons R. Discover Denver. The World Class City—City Planning and Reinvestment, 1983-2014. 2016. [Google Scholar]
68.Robinson D, Rhodes D, Van Sluytman L. Watching neighborhoods vanish: The intertwining of gentrification, race, class, and policy. J Poverty. 2020;24(5–6):473–92. [Google Scholar]
69.Prener CG. Demographic change, segregation, and the emergence of peripheral spaces in St. Louis, Missouri. Appl Geogr. 2021;133:102472. doi: 10.1016/j.apgeog.2021.102472 [DOI] [Google Scholar]
70.White AG, Guikema SD, Logan TM. Urban population characteristics and their correlation with historic discriminatory housing practices. Appl Geogr. 2021;132:102445. doi: 10.1016/j.apgeog.2021.102445 [DOI] [Google Scholar]
71.Goldblatt R, Omer I. The association between land-use distribution and residential patterns: the case of mixed Arab-Jewish cities in Israel. J Urban Regional Anal. 2020;6(1): 15-34. doi: 10.37043/JURA.2014.6.1.2 [DOI] [Google Scholar]

PLoS One. 2025 Mar 3;20(3):e0317988. doi: 10.1371/journal.pone.0317988.r001

Author response to Decision Letter 0

8 Feb 2024

PLoS One. doi: 10.1371/journal.pone.0317988.r002

Decision Letter 0

Christopher A Lepczyk

3 Sep 2024

PONE-D-24-05471Land use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city.PLOS ONE

Dear Dr. Ibsen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. These comments are located at the bottom of this message.

Please submit your revised manuscript by Oct 18 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Christopher A. Lepczyk

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please note that PLOS ONE has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, all author-generated code must be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse.

3. Thank you for stating the following in your Competing Interests section:

"NO"

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

4. We note that your Data Availability Statement is currently as follows: [All relevant data are within the manuscript and its Supporting Information files]

Please confirm at this time whether or not your submission contains all raw data required to replicate the results of your study. Authors must share the “minimal data set” for their submission. PLOS defines the minimal data set to consist of the data required to replicate all study findings reported in the article, as well as related metadata and methods (https://journals.plos.org/plosone/s/data-availability#loc-minimal-data-set-definition).

For example, authors should submit the following data:

- The values behind the means, standard deviations and other measures reported;

- The values used to build graphs;

- The points extracted from images for analysis.

Authors do not need to submit their entire data set if only a portion of the data was used in the reported study.

If your submission does not contain these data, please either upload them as Supporting Information files or deposit them to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of recommended repositories, please see https://journals.plos.org/plosone/s/recommended-repositories.

If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent. If data are owned by a third party, please indicate how others may request data access.

5. We note that [Figure 1] in your submission contain [satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

6. Please include a new copy of Table xxxx in your manuscript; the current table is difficult to read. Please follow the link for more information: https://blogs.plos.org/plos/2019/06/looking-good-tips-for-creating-your-plos-figures-graphics/

Additional Editor Comments:

Associate Editor:

General Comments

My apologies for the long delay in returning a decision on the manuscript. We went through an excessively large number of reviewers to obtain two qualified individuals to assess the work. As you will note, both reviewers found merit in the work and appreciated the level of analysis done. The main item from the reviews that warrants further consideration is the use/inclusion of zoning data. If zoning data are available for part or all of the time series, it would greatly aid your analyses and should be pursued. If these data are unavailable, then simply note that in a rebuttal. Finally, I would suggest including a paragraph in the Discussion that clearly describes all the limitations of the work.

Specific Comments

Abstract. Include main research questions/hypotheses.

L42. Change greenspace to green space here and throughout the ms.

L46. One point that has come up that may want to note is that redlining was not the singular factor that has been recently described, but one of several that may co-occur, etc. You mostly note this point already, but may touch upon a bit more that it isn’t just a simple classification that explains the findings of other studies. Likewise, this point also relates to the reviewer’s comment about zoning.

L101-129. Much of this text can be moved to Methods or deleted. L101 should start the final paragraph that lays out the overarching goal/question followed by the specific questions/hypotheses/objectives.

L130. I would suggest revising this first sentence a bit to indicate the gap in knowledge that you are seeking to answer.

L174. Please note here or elsewhere in the Methods what datum and projection were used in your geographic analyses.

L192. How are you dealing with different resolutions of the datasets?

L215. Revise to ‘This approach…’

L223. Did you evaluate collinearity amongst terms first? If not, please do. Also, did you have to transform any of the data? Finally, are you just using p-values here or AICc?

L241. Here or elsewhere in Methods indicate the statistical software you used for analyses.

L244. I would suggest cutting this sentence. Results here should be very succinct. You found x relationship, with y parameters being important.

Fig. 2 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category and years.’ Also, no description of panel B is noted.

L276. Can revise to just indicate result, do not need to describe the statistical approach used as that was presented in the Methods.

L277. Provide the full statistical results (F = xxx, DF = x,y, p = z.zz) here and elsewhere in Results. Also, delete the sentence that starts at this line as you already explained post hoc approach in Methods.

L290. See previous comment on including detailed stats.

L301, L315, L358. Provide the statistical details of the analyses here.

Fig. 3 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category and years.’

Fig. 4 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category, income class, and years.’

L354. Move this item to Methods.

Fig. 5 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category and years.’ And ‘Regression line equations and p values by HOLC category are provided.’

L374. Can cut most of first two sentences. Begin with answers to your questions.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The principal claim of this article is succinctly summarized in the title: “Land use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city,” but the study asks and answers much more far-reaching claims than this. And, the study effectively qualifies the specific role of “redlining,” meaning the impact of the 1938 HOLC Security Maps as a causal factor in this overall process. Instead, the authors use the HOLC categories as a snapshot of the land use and demographic conditions at the starting-point in their longitudinal study. The goal at the heart of the study is to track long-term changes in land use diversity and demographic areal composition from 1937 to 2019. While the long-term correlations between HOLC categories (developed originally and used by lending agencies thereafter for the purpose of qualifying applicants for mortgages) and downstream social inequality, urban amenities, environmental injustice, and other factors has been well-documented now in numerous studies amply cited in this study, the key innovation here is the use of an allegedly unique—for Denver-- high-resolution land-use data series (Drummond et al 2019) , for Denver in the years 1937, 1957, 1977, and 1997. This data series allows the authors to a) set a benchmark of land use prior to any possible impact of the HOLC maps, and b) track changes in 20-year intervals between the onset of the redlining era and pour contemporary, particularly the first two decades of the 21st century.

In the authors’ own words, “The spatial patterns of this connection between the built environment, past discriminatory policies, and the potential vulnerability to neighborhood gentrification have been connected from the present to the HOLC boundaries (30), but to understand how these spatial patterns emerged over time, time series data is needed. (96-101). The authors address three specific questions: “1. How have patterns in land use and demographics developed and diverged in formally categorized HOLC areas across Denver from 1937 to 2019? 2. How has the diversity of demography in Denver neighborhoods connected to HOLC areas changed over the same period? 3. How have patterns of inequality in land use and demographics from the previous decades influenced rapid present-day changes in Denver’s demographics and resulting neighborhood-specific patterns in gentrification?” (129-138).

Overall the authors are successful in answering all three of these questions, but there is a major factor that they both recognized and fail to include in their analyses, which calls into question the whole rationale for depending on HOLC classifications in the first place: Denver’s zoning ordinances.

The authors do show statistically significant patterns in which the land-use diversity was reinforced in HOLC category D areas across the entire time period, which residential, single-family land use also increased in areas A and B. Proportions of college-educated, white, and higher-income households were all significantly lower for HOLD D areas than for HOLC-A areas over time. Given the goals of the study, a significant achievement is that the authors are able to show significant changes within the overall trends from 1937-2019 in two time-periods: 1) a significant impact of urban renewal policies and projects during the 1960-1980 interval, and 2) the impact of gentrification in areas previously disadvantaged by the longer-term trends, in the early 20th century period.

While all of this is very solid statistically, this reviewer believes that the authors made a critical methodological mistake by not identifying and including the geographic boundaries of the city’s zoning ordinances, along with the land-use restrictions codified by those ordinances for those geographic boundaries, as independent variables in addition of the HOLC boundaries. The HOLC boundaries, after all, were established for the single purpose of evaluating the loan-security risk of residential properties. Thereafter, they were used or adapted by banks for decades, but only for the purpose of home loan qualifications. The principal innovation of this study is to use the Drummond, et al land-use maps, which capture all categories of land use. The authors use these maps primarily along the dimension of “diversity,” which is essentially bi-variate in their usage: they seek to distinguish high-diversity areas (where industrial and other non-residential uses are concentrated), and low-diversity, where primarily residential areas are concentrated. Zoning ordinances were invented in the early 20th century to exclude industrial and other commercial uses from certain residential areas. (and also to optimize industrial development). They did not exclude residential development within industrial/commercial areas, but they did exclude industrial/commercial development in areas newly developed to be exclusively residential. This all happened concurrent with the rise of racial restrictions, usually to exclude non-whites from those same all-residentially zoned areas. While this reviewer is not familiar with Denver’s specific zoning ordinances, based on their use in cities with which the reviewer does have familiarity, it is nearly certain that they were used to exclude land use “diversity” from residential areas over this long period. It is also likely that they were revised significantly over the years. It is also certain that zoning ordinances have a direct impact on land use, while HOLC categories have only an indirect impact. And, it is very well known that residential property values are directly impacted by proximity to nuisances typical of industrial and commercial land uses. It is likely, therefore, that the exclusivity of HOLC areas A and B, and the land use diversity of areas C and D, can be explained entirely by the regulatory effect of these ordinances. The word “zoning” only appears three times in this manuscript (95, 403, 416), but in each of these passages the authors acknowledge the importance of zoning ordinances.

The authors laudably avoid assigning HOLC maps a singular causal role in the longer temporal patterns of land use and demographic spatial inequality. They show empirically that the inequalities of land use and demographic geography pre-existed the HOLC maps, and that changes occurred long after their implementation. This reviewer believes that the failure to consider the role of zoning geographies across this period is a major oversight. While the HOLC maps have come to stand for a long-term inscription of inequality in urban landscapes, the authors needn’t have tied their study so closely to those maps which, as they acknowledge, do not even cover downtown areas. The status quo of land use and social demography in 1937 and subsequent years could have been classified by an independent set of area units, and tracked over time for long-term patterns. The use of zoning geographies would have gone a long way toward helping to explain those patterns, as city planning authorities were required to certify all building project according to these maps—and not to the HOLC maps. This reviewer does not recommend removing the HOLC geographies from the study—far from it: the inclusion of the HOLC geographies and categories should still be an important set of variables in the study. But they should be used in conjunction with the explanatory power of the zoning geographies. This approach will a) give a much more accurate view of the long-term urban developmental historical process (recognizing the factor most likely to shape land use), and b) help to advance the latest state of the art in understanding the impact of the HOLC maps. My recommendation, therefore, does not imply scrapping the entire approach of this study. Instead, it suggests the addition of mapping layers with associated attributes, most probably chosen to coincide with the four land-use data points of 1937, 57, 77, and 97, and re-running their analyses to include the independent variable of land use restrictions on every space in the study.

Reviewer #2: This research analyzed changes in land use and socioeconomic factors, considering the impacts of historical redlining, to identify current patterns of gentrification in Denver, USA. They used boxplots and statistical analysis to test three research questions developed in this research. Given the increasing academic interest in the lingering impacts of redlining on urban dynamics, the findings here are noteworthy and provide some implications. Before publication, the manuscript needs to address the following issues.

Major issue

Introduction

1. Although the authors noted some critiques of using HOLC data in the first paragraph (e.g., over-causality), it would be beneficial to also address another criticism, such as the skepticism regarding the actual impact of HOLC (check this:

https://doi.org/10.1177/0096144203029004002).

2. More recently published papers should be cited in lines 88-90.

https://doi.org/10.1016/j.landurbplan.2024.105019

https://doi.org/10.1016/j.landurbplan.2024.105028

https://doi.org/10.1080/13549839.2024.2380854

3. The first and second research questions should be revised for clearer distinction. Currently, they have some overlap.

Methods

1. The authors used the Shannon-Wiener index (H') to calculate the compositional features of land use. Are there other metrics or methods that can assess the configurational aspect of land use? Even if two districts have similar compositional attributes, their configurational characteristics may differ significantly, and these can provide some information.

2. Why do the authors use a quasi-binomial distribution to fit their models? Providing a rationale for this choice would strengthen the paper.

3. I wonder whether the gentrification index was developed by the authors or adapted from other studies. If the authors developed it, why was "the bottom 40th percentile" chosen as a criterion? In addition, the phrase "1 of the gentrification metric (in line 237)" is confusing: so tracts eligible for gentrification satisfy at least one of the four criteria or only the first criterion?

Results

1. The findings described in lines 349-352 were difficult to verify through Fig. 5A. I recommend revising both the figure and the manuscript.

2. How many observations were used to estimate each statistical model in Fig. 5b (supplementary figure also)? Very few samples appear in each HOLC category, making it difficult to calculate statistical significance.

Discussion

1. More references should be used to compare or support this research's findings. Papers addressing regions beyond Denver could offer valuable implications.

2. Although some limitations are mentioned at the end of the Discussion, additional methodological and conceptual limitations should be provided.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy .

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/ . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org . Please note that Supporting Information files do not need this step.

PLoS One. 2025 Mar 3;20(3):e0317988. doi: 10.1371/journal.pone.0317988.r003

Author response to Decision Letter 1

13 Nov 2024

AE comments:

Specific Comments

Abstract. Include main research questions/hypotheses.

L42. Change greenspace to green space here and throughout the ms.

Corrected

We have added a section focused on how zoning also plays a role in determining both the trend of homeownership and demographics, as well as the potential for HOLC C and D areas to be prime candidates for gentrification. However, as we explain in this section and in the associated figures, unfortunately the zoning data that is available is not usable to create a legitimate time-series. Additionally, we provide our reasons for using HOLC categories due to the fact they offer a standardized description of the various categories prior to subsequent lending impacts, and their fixed spatial extents provide a consistent location to analyze change over time. We do feel that by adding this section and highlighting that there is a greater flexibility in land use regulations in redlined districts, and that redlined districts are more vulnerable to shifts in zoning because they have fewer areas zoned single-family residential better make the case the HOLC boundaries were not a singular factor in driving land use and demographic trends, but are strongly linked to other regulatory processes.

We have incorporated this section into the Methods sub-section “Analysis of land-use diversity”, and removed any redundant text.

L130. I would suggest revising this first sentence a bit to indicate the gap in knowledge that you are seeking to answer.

The paragraph header has been adjusted to bring the main gaps in knowledge (the lack of time-series data, the importance of considering both land use and demographics) to the forefront before introducing our research questions.

L174. Please note here or elsewhere in the Methods what datum and projection were used in your geographic analyses.

Corrected

L192. How are you dealing with different resolutions of the datasets?

The resolution of the Drummond and Chen data sets are the same (1 m). We have edited the text to ensure that point is clear.

L215. Revise to ‘This approach…’

Corrected

L223. Did you evaluate collinearity amongst terms first? If not, please do. Also, did you have to transform any of the data? Finally, are you just using p-values here or AICc?

We have added text explaining how multicollinearity was expected between factors of race, education, and income. So we all demographic models were done independent from each other to isolate the specific demographic patterns.

L241. Here or elsewhere in Methods indicate the statistical software you used for analyses.

Corrected

L244. I would suggest cutting this sentence. Results here should be very succinct. You found x relationship, with y parameters being important.

Corrected

Fig. 2 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category and years.’ Also, no description of panel B is noted.

Corrected.

L276. Can revise to just indicate result, do not need to describe the statistical approach used as that was presented in the Methods.

We have removed superfluous text which was already described in the Methods section

Corrected and appreciated! On producing the statistics the authors discovered a typo in the statistical code, when corrected we found that HOLC D was significantly different in Shannon Diversity back to 1937. This actually bolsters our argument that redlining practices were not the creative force for present day land use segregation, but did correlate with increases in segregation of land use diversity in the following decades. We have corrected this text here and in the discussion.

L290. See previous comment on including detailed stats.

L301, L315, L358. Provide the statistical details of the analyses here.

We have added statistical details for all linearly modeled relationships to the text

Fig. 3 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category and years.’

Corrected

Fig. 4 legend. Delete ‘We determined significance through a generalized linear model with interactions between HOLC category, income class, and years.’

Corrected

L354. Move this item to Methods.

We have moved to the methods and added text on how we conducted the analysis.

Corrected

L374. Can cut most of first two sentences. Begin with answers to your questions.

Corrected

Reviewer 1 Comments:

We greatly thank Reviewer 1 for their in-depth consideration of this work. We also completely agree with them! Previously we had not included any analysis on zoning (aside from the comments we had already made expressing that zoning ordination would also be considered a driving force behind patterns in land use configuration and diversity. This was borne out of a limited access to historical zoning data. However, thanks to this comment we have engaged with the GIS office of the City of Denver and received some historical zoning data. However, this data only goes back to 1955, and only indicated the year when a zoning ordinance was changed, but it does not indicate exactly what the previous ordinance was. Thus, we cannot exactly track a similar time-series of zoning influence on land use as we do with the rest of the analysis.

However, we have added a section focused on how zoning also plays a role in determining both the trend of homeownership and demographics, as well as the potential for HOLC C and D areas to be prime candidates for gentrification. However, as we explain in this section and in the associated figures, unfortunately the zoning data that is available is not usable to create a legitimate time-series. Additionally, we provide our reasons for using HOLC categories due to the fact they offer a standardized description of the various categories prior to subsequent lending impacts, and their fixed spatial extents provide a consistent location to analyze change over time.

With the limited zoning data we have we were able to present some analysis that reflects similar patterns to what we have shown in regards to land cover change. Firstly, we have found that former HOLC D areas currently are zoned to include a greater diversity of zoning ordinances, and moreover, by tracking the dates of zoning change, we found HOLC C and D areas experienced more regular changes in their zoning ordinances over a period of 1955-2020. These results are now included as Supplemental Figure 2.

We do feel that by adding this addition text, analysis, and highlighting that there is a greater flexibility in land use regulations in redlined districts, and that redlined districts are more vulnerable to shifts in zoning because they have fewer areas zoned single-family residential better make the case the HOLC boundaries were not a singular factor in driving land use and demographic trends, but are strongly linked to other regulatory processes.

Reviewer 2 Comments:

Attachment

Submitted filename: PLOS_reviewer_comments_revised.docx

pone.0317988.s003.docx^{(31.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0317988.r004

Decision Letter 1

Christopher A Lepczyk

9 Jan 2025

Land use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city.

PONE-D-24-05471R1

Dear Dr. Ibsen,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Both reviewers and I appreciate all the revisions to the text and believe this will be a valuable addition to the literature.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Christopher A. Lepczyk

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: I am very happy with the authors´ response to my recommendation that they include historical Denver zoning data in their analysis. They have incorporated a fresh analysis of the zoning data in a convincing way that reinforces their previous analysis and conclusions. Note a very tiny detail, a typo on line 805 describing supplemental figure 2, reads "differences," should be "different."

Reviewer #2: Thank you to the authors for their efforts in addressing my comments and suggestions. I believe the revised version has reached a publishable level.

However, I have a few minor points to mention, which I believe can be addressed during the galley proof:

1. In Fig. 2, the sub-label (a) is missing in the current version.

2. Some paragraphs, such as those on pages 24 and 25, are too long. I recommend dividing these paragraphs to enhance readability.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean? ). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy .

Reviewer #1: No

Reviewer #2: No

**********

PLoS One. doi: 10.1371/journal.pone.0317988.r005

Acceptance letter

Christopher A Lepczyk

PONE-D-24-05471R1

PLOS ONE

Dear Dr. Ibsen,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Christopher A. Lepczyk

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

(TIF)

pone.0317988.s001.tif^{(971.9KB, tif)}

S2 Fig. The influence of zoning ordinances, and their changes on land use configuration across former Home Owners’ Loan Corporation (HOLC) categories in Denver, Colorado.

(TIF)

pone.0317988.s002.tif^{(1MB, tif)}

Attachment

Submitted filename: PLOS_reviewer_comments_revised.docx

pone.0317988.s003.docx^{(31.8KB, docx)}

Data Availability Statement

[pone.0317988.ref001] 1.Nelson RK, Winling L. Mapping Inequality: Redlining in New Deal America [Internet]. 2023 [cited 2022 Dec 12]. Available from: https://dsl.richmond.edu/panorama/redlining

[pone.0317988.ref002] 2.Huang SJ, Sehgal NJ. Association of historic redlining and present-day health in Baltimore. Liu SY, editor. PLoS ONE. 2022. Jan 19;17(1):e0261028. doi: 10.1371/journal.pone.0261028 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref003] 3.Locke DH, Hall B, Grove JM, Pickett STA, Ogden LA, Aoki C, et al. Residential housing segregation and urban tree canopy in 37 US cities. npj Urban Sustainability. 2021;1(1):15. [Google Scholar]

[pone.0317988.ref004] 4.Nardone AL, Casey JA, Rudolph KE, Karasek D, Mujahid M, Morello-Frosch R, et al. Associations between historical redlining and birth outcomes from 2006 through 2015 in California. Khan HTA, editor. PLoS ONE. 2020. Aug 7;15(8):e0237241. doi: 10.1371/journal.pone.0237241 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref005] 5.Lane H, Morello-Frosch R, Marshall J, Apte J. Historical redlining is associated with present-day air pollution disparities in U.S. cities. Environ Sci Technol Lett. 2022;9(4):345–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref006] 6.Moxley D, Fischer B. Historic HOLC redlining in Indianapolis and the legacy of environmental impacts. Journal of Public and Environmental Affairs [Internet]. 2020. [cited 2024 Jan 29];1(1). Available from: https://scholarworks.iu.edu/journals/index.php/jpea/article/view/30321 [Google Scholar]

[pone.0317988.ref007] 7.Fishback P, Rose J, Snowden KA, Storrs T. New evidence on redlining by federal housing programs in the 1930s. J Urban Eco. 2022;103:462. [Google Scholar]

[pone.0317988.ref008] 8.Markley S. Federal ‘redlining’ maps: a critical reappraisal. Urban Stud. 2023. doi: 10.1177/00420980231182336 [DOI] [Google Scholar]

[pone.0317988.ref009] 9.Drummond MA, Stier MP, Diffendorfer J (Jay) E. Historical land use and land cover for assessing the northern colorado front range urban landscape. J Maps. 2019;15(2):89–93. doi: 10.1080/17445647.2018.1548383 [DOI] [Google Scholar]

[pone.0317988.ref010] 10.Ruggles S, Flood S, Sobek M, Backman D, Chen A, Coope G, et al. IPUMS USA: Version 15.0 [Internet]. Minneapolis, MN; 2024. Available from: doi: 10.18128/D010.V15.0 [DOI] [Google Scholar]

[pone.0317988.ref011] 11.Mauck LM. Five points neighborhood of Denver. Arcadia Publishing; 2001. [Google Scholar]

[pone.0317988.ref012] 12.Verell BD. Spring valley, Washington DC: Changing land use and demographics from 1900-2000. Geographical Bulletin - Gamma Theta Upsilon. 2008;49(2):103–19. [Google Scholar]

[pone.0317988.ref013] 13.Hillier AE. Redlining and the home owners’ loan corporation. J Urban Hist. 2003;29(4):394–420. doi: 10.1177/0096144203029004002 [DOI] [Google Scholar]

[pone.0317988.ref014] 14.Hyra DS. Conceptualizing the new urban renewal. Urban Aff Review. 2012;48(4):498–527. doi: 10.1177/1078087411434905 [DOI] [Google Scholar]

[pone.0317988.ref015] 15.Teaford JC. Urban renewal and its aftermath. Housing Policy Debate. 2000;11(2):443–65. [Google Scholar]

[pone.0317988.ref016] 16.Tiagi R. Occupational mobility of black migrants in the West during the 1950s. Labor His. 2015;56(1):58–68. doi: 10.1080/0023656x.2015.991563 [DOI] [Google Scholar]

[pone.0317988.ref017] 17.Grove M, Ogden L, Pickett S, Boone C, Buckley G, Locke DH, et al. the legacy effect: understanding how segregation and environmental injustice unfold over time in baltimore. Ann Am As Geogr. 2017;108(2):524–37. doi: 10.1080/24694452.2017.1365585 [DOI] [Google Scholar]

[pone.0317988.ref018] 18.Horr A, Hunkler C, Kroneberg C. Ethnic discrimination in the german housing market. Zeitschrift für Soziologie. 2018;47(2):134–46. doi: 10.1515/zfsoz-2018-1009 [DOI] [Google Scholar]

[pone.0317988.ref019] 19.Aalbers MB. Place-based and race-based exclusion from mortgage loans: evidence from three cities in the Netherlands. J Urban Aff. 2007;29(1):1–29. doi: 10.1111/j.1467-9906.2007.00320.x [DOI] [Google Scholar]

[pone.0317988.ref020] 20.Haferburg C, Huchzermeyer M. Redlining or renewal? The space-based construction of decay and its contestation through local agency in Brixton, Johannesburg. In: Kirkness P, Tije-Dra A, editors. Negative Neighbourhood Reputation and Place Attachment: The Production and Contestation of Territorial Stigma [Internet]. 1st ed. Routledge; 2017. [cited 2023 Sep 24]. Available from: https://www.taylorfrancis.com/books/9781317089537 [Google Scholar]

[pone.0317988.ref021] 21.Wilson B. Urban Heat Management and the Legacy of Redlining. J Am Plan As. 2020;86(4):443–57. doi: 10.1080/01944363.2020.1759127 [DOI] [Google Scholar]

[pone.0317988.ref022] 22.Jung MC, Yost MG, Dannenberg AL, Dyson K, Alberti M. Legacies of redlining lead to unequal cooling effects of urban tree canopy. Landscape Urban Plan. 2024;246:105028. doi: 10.1016/j.landurbplan.2024.105028 [DOI] [Google Scholar]

[pone.0317988.ref023] 23.Walker RH, Keeler BL, Derickson KD. The impacts of racially discriminatory housing policies on the distribution of intra-urban heat and tree canopy: a comparison of racial covenants and redlining in Minneapolis, MN. Landscape Urban Plan. 2024;245:105019. doi: 10.1016/j.landurbplan.2024.105019 [DOI] [Google Scholar]

[pone.0317988.ref024] 24.Donovan GH, Prestemon JP, Butry DT, Kaminski AR, Monleon VJ. The politics of urban trees: tree planting is associated with gentrification in Portland, Oregon. For Policy Econ. 2021;124:10.1016/j.forpol.2020.102387. doi: 10.1016/j.forpol.2020.102387 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref025] 25.Kolko J. The Determinants of Gentrification. SSRN Electronic Journal [Internet]. 2007. Dec. Available from: http://www.ssrn.com/abstract=985714 [Google Scholar]

[pone.0317988.ref026] 26.Anguelovski I, Connolly JJT, Cole H, Garcia-Lamarca M, Triguero-Mas M, Baró F, et al. Green gentrification in European and North American cities. Nat Commun. 2022;13(1):3816. doi: 10.1038/s41467-022-31572-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref027] 27.Atkinson R, Wulff M, Reynolds M, Spinney A. Gentrification and displacement: the household impacts of neighbourhood change. Aust Housing Urban Res Inst. 2011. [Google Scholar]

[pone.0317988.ref028] 28.Leguizamon S, Christafore D. The influence of land use regulation on the probability that low-income neighbourhoods will gentrify. Urban Stud. 2020;58(5):993–1013. doi: 10.1177/0042098020940163 [DOI] [Google Scholar]

[pone.0317988.ref029] 29.Curran W. From the frying pan to the oven: Gentrification and the experience of industrial displacement in Williamsburg, Brooklyn. Urban Stud. 2007;44(8):1427–40. doi: 10.1080/00420980701373438 [DOI] [Google Scholar]

[pone.0317988.ref030] 30.Davis J. How do upzonings impact neighborhood demographic change? examining the link between land use policy and gentrification in New York City. Land Use Policy. 2021;103:105347. doi: 10.1016/j.landusepol.2021.105347 [DOI] [Google Scholar]

[pone.0317988.ref031] 31.Gould KA, Lewis TL. Green gentrification: Urban sustainability and the struggle for environmental justice. Routledge. 2016;192. [Google Scholar]

[pone.0317988.ref032] 32.Pham T-T-H, Apparicio P, Landry S, Lewnard J. Disentangling the effects of urban form and socio-demographic context on street tree cover: a multi-level analysis from Montréal. Landscape Urban Plan. 2017;157:422–33. doi: 10.1016/j.landurbplan.2016.09.001 [DOI] [Google Scholar]

[pone.0317988.ref033] 33.Gibbons J. Examining the long-term influence of New Deal era redlining on contemporary gentrification. Urban Stud. 2023;60(14):2816–34. [Google Scholar]

[pone.0317988.ref034] 34.Lazin FA. The housing policies for Ethiopian immigrants in Israel: Spatial segregation, economic feasbility and political acceptability. Nationalism Ethn Polit. 1997;3(4):39–68. doi: 10.1080/13537119708428517 [DOI] [Google Scholar]

[pone.0317988.ref035] 35.Singh G, Vithayathil T, Pradhan KC. Recasting inequality: residential segregation by caste over time in urban India. Environ Urban. 2019;31(2):615–34. doi: 10.1177/0956247818812330 [DOI] [Google Scholar]

[pone.0317988.ref036] 36.U.S. Census Bureau. Southern and Western Regions Experienced Rapid Growth This Decade [Internet]. United States Census; 2020. Available from: https://www.census.gov/newsroom/press-releases/2020/south-west-fastest-growing.html

[pone.0317988.ref037] 37.Bunyak D, Simmons TH, Simmons RL. Denver area post-world war II suburbs. Colorado Dep Trans. 2011;288. [Google Scholar]

[pone.0317988.ref038] 38.Abbott C. Suburb and city: changing patterns of socioeconomic status in metropolitan Denver since 1940. Soc Sci Hist. 1977;2(1):53. [Google Scholar]

[pone.0317988.ref039] 39.Connor D, Gutmann M, Cunningham A, Clement K, Leyk S. How entrenched is the spatial structure of inequality in cities? evidence from the integration of census and housing data for Denver from 1940 to 2016. Ann Am Ass Geogr. 2020;110(4):1022–39. [Google Scholar]

[pone.0317988.ref040] 40.Denver Urban Renewal Authority. 50 years of revitalizing Denver [Internet]. 2008. Available from: https://digital.denverlibrary.org/nodes/view/822540

[pone.0317988.ref041] 41.Avila E, Rose MH. Race, culture, politics, and urban renewal: an introduction. J Urban Hist. 2009;35(3):335–47. [Google Scholar]

[pone.0317988.ref042] 42.Frey WH. America’s Largest Metro Areas [Internet]. Brookings Mountain West; 2022. p. 1–32. Available from: https://digitalscholarship.unlv.edu/brookings_policybriefs_reports/11 [Google Scholar]

[pone.0317988.ref043] 43.Shannon CE. A mathematical theory of communication. Bell Syst Techn J. 1948;27(3):53. [Google Scholar]

[pone.0317988.ref044] 44.Liu H, Weng Q. Landscape metrics for analysing urbanization-induced land use and land cover changes. Geocarto Int. 2013;28(7):582–93. doi: 10.1080/10106049.2012.752530 [DOI] [Google Scholar]

[pone.0317988.ref045] 45.Sapena M, Ruiz LÁ. Analysis of land use/land cover spatio-temporal metrics and population dynamics for urban growth characterization. Comput, Environ Urban Syst. 2019;73:27–39. doi: 10.1016/j.compenvurbsys.2018.08.001 [DOI] [Google Scholar]

[pone.0317988.ref046] 46.Alsavada MO, Karimi K. The Spatio-Functional Role of Navigable Urban Canals in the City: Cases From London and Amsterdam. UP [Internet]. 2023 Aug 17. [cited 2024 Sep 24];8(3). Available from: https://www.cogitatiopress.com/urbanplanning/article/view/6740 [Google Scholar]

[pone.0317988.ref047] 47.Yoshida T, Tanaka K. Land-use diversity index: a new means of detecting diversity at landscape level. Landscape Ecol Eng. 2005;1(2):201–6. doi: 10.1007/s11355-005-0022-0 [DOI] [Google Scholar]

[pone.0317988.ref048] 48.Hoffman J, Shandas V, Pendleton N. The effects of historical housing policies on resident exposure to intra-urban heat: a study of 108 US urban areas. Climate. 2020;8(1):12. [Google Scholar]

[pone.0317988.ref049] 49.Chen B, Tu Y, Song Y, Theobald DM, Zhang T, Ren Z, et al. Mapping essential urban land use categories with open big data: Results for five metropolitan areas in the United States of America. ISPRS J Photogramm Remote Sens. 2021;178:203–18. doi: 10.1016/j.isprsjprs.2021.06.010 [DOI] [Google Scholar]

[pone.0317988.ref050] 50.U.S. Census Bureau’s American Community Survey Office. American Fact Finder - 2019 American Community Survey. 2019. US Census Bureau. [Google Scholar]

[pone.0317988.ref051] 51.Fry R, Taylor P. The rise of residential segregation by income. Washington, DC: Pew Research Center. 2012;(202):26. [Google Scholar]

[pone.0317988.ref052] 52.Freeman L. Displacement or succession?: residential mobility in gentrifying neighborhoods. Urban Affair Rev. 2005;40(4):463–91. [Google Scholar]

[pone.0317988.ref053] 53.Maciag Michael. Gentrification Report Methodology. Governing: The Future of States and Localities [Internet]. 2015. Jan 23; Available from: https://www.governing.com/archive/gentrification-report-methodology.html [Google Scholar]

[pone.0317988.ref054] 54.Logan JR, Stults BJ, Xu Z. Validating population estimates for harmonized census tract data, 2000–2010. Ann Am Ass Geogr. 2016. Sep 2;106(5):1013–29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref055] 55.Vo, Thuy L. NYC 311 complaints and demographic analysis — 2010 to 2018 [Internet]. 2018. Available from: https://github.com/BuzzFeedNews/2018-06-nyc-311-complaints-and-gentrification

[pone.0317988.ref056] 56.RStudio Team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: RStudio, PBC; 2021. Available from: http://www.rstudio.com/ [Google Scholar]

[pone.0317988.ref057] 57.Lens MC. Zoning, land use, and the reproduction of urban inequality. Annu Rev Sociol. 2022;48(1):421–39. doi: 10.1146/annurev-soc-030420-122027 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0317988.ref058] 58.Baics G, Meisterlin L. Zoning before zoning: land use and density in mid-nineteenth-century New York City. Ann Am Assoc Geogr. 2016;106(5):1152–75. [Google Scholar]

[pone.0317988.ref059] 59.Irwin EG, Geoghegan J. Theory, data, methods: developing spatially explicit economic models of land use change. Agricult, Ecosyst Environ. 2001;85(1):7–24. [Google Scholar]

[pone.0317988.ref060] 60.Ellickson RC. The zoning straitjacket: the freezing of American neighborhoods of single-family houses. Indiana Law J. 2020;96:395. [Google Scholar]

[pone.0317988.ref061] 61.Brooks L, Schuetz J. Does housing growth in Washington, D.C., reflect land use policy changes?. Cityscape. 2023;25(2):203–24. [Google Scholar]

[pone.0317988.ref062] 62.Haurin DR, Dietz RD, Weinberg BA. The impact of neighborhood homeownership rates: a review of the theoretical and empirical literature. J Housing Res. 2002;13(2):119–51. [Google Scholar]

[pone.0317988.ref063] 63.Been V, Madar J, McDonnell S. Urban land‐use regulation: are homevoters overtaking the growth machine?. J Empir Leg Stud. 2014;11(2):227–65. [Google Scholar]

[pone.0317988.ref064] 64.Page B, Ross E. Envisioning the urban past: GIS reconstruction of a lost Denver district. Front Digit Human. 2015;2. [Google Scholar]

[pone.0317988.ref065] 65.Page B, Ross E. Legacies of a contested campus: urban renewal, community resistance, and the origins of gentrification in Denver. Urban Geogr. 2016;38(9):1293–328. doi: 10.1080/02723638.2016.1228420 [DOI] [Google Scholar]

[pone.0317988.ref066] 66.Engels B. Capital flows, redlining and gentrification: the pattern of mortgage lending and social change in glebe, Sydney, 1960–1984*. Int J Urban Regional Res. 1994;18(4):628–57. doi: 10.1111/j.1468-2427.1994.tb00290.x [DOI] [Google Scholar]

[pone.0317988.ref067] 67.Simmons T, Simmons R. Discover Denver. The World Class City—City Planning and Reinvestment, 1983-2014. 2016. [Google Scholar]

[pone.0317988.ref068] 68.Robinson D, Rhodes D, Van Sluytman L. Watching neighborhoods vanish: The intertwining of gentrification, race, class, and policy. J Poverty. 2020;24(5–6):473–92. [Google Scholar]

[pone.0317988.ref069] 69.Prener CG. Demographic change, segregation, and the emergence of peripheral spaces in St. Louis, Missouri. Appl Geogr. 2021;133:102472. doi: 10.1016/j.apgeog.2021.102472 [DOI] [Google Scholar]

[pone.0317988.ref070] 70.White AG, Guikema SD, Logan TM. Urban population characteristics and their correlation with historic discriminatory housing practices. Appl Geogr. 2021;132:102445. doi: 10.1016/j.apgeog.2021.102445 [DOI] [Google Scholar]

[pone.0317988.ref071] 71.Goldblatt R, Omer I. The association between land-use distribution and residential patterns: the case of mixed Arab-Jewish cities in Israel. J Urban Regional Anal. 2020;6(1): 15-34. doi: 10.37043/JURA.2014.6.1.2 [DOI] [Google Scholar]

PERMALINK

Land-use and socioeconomic time-series reveal legacy of redlining on present-day gentrification within a growing United States city

Peter C Ibsen

Anna Bierbrauer

Lucila M Corro

Zachary H Ancona

Mark Drummond

Kenneth J Bagstad

Jay E Diffendorfer

Roles

Abstract

Introduction

Methods

Study area

Analysis of land-use diversity

Fig 1. High-resolution land-use data of Denver from 1937–1997.

Analysis of demography time series

Analysis of gentrification

Results

Land-use diversity

Fig. 2. Proportions of land-use classes in Home Owners’ Loan Corporation (HOLC) categories and their diversity across time.

Table 1. Land use percentages aggregated to Home Owners’ Loan Corporation (HOLC) categories from 1937-1997 in Denver Colorado.

Demographics

Fig. 3. Patterns of racial and educational demographics of Denver residents within Home Owners’ Loan Corporation (HOLC) categories over eight decades.

Table 2. Percent changes to demographic metrics in Home Owners’ Loan Corporation (HOLC) categories. Education and racial data start in 1940 and income data start in 1960.

Fig. 4. Patterns of income of Denver residents within Home Owners’ Loan Corporation (HOLC) categories over eight decades.

Gentrification

Fig. 5. Gentrification metrics for Home Owners’ Loan Corporation (HOLC) category and relationship to race.

Discussion

Drivers of land-use diversity

Shifts in land-use patterns and demographics

Land-use and demographic drivers of gentrification

Limitations

Conclusion

Disclaimer

Supporting information

Data Availability

Funding Statement

References

Author response to Decision Letter 0

Decision Letter 0

Christopher A Lepczyk

Roles

Author response to Decision Letter 1

Decision Letter 1

Christopher A Lepczyk

Roles

Acceptance letter

Christopher A Lepczyk

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases