Navigation – Plan du site

AccueilNuméros4Segregation and environmental ine...

Segregation and environmental inequalities in French cities

Ségrégation et inégalités environnementales dans les villes françaises
Mihai Tivadar et Yves Schaeffer

Résumés

La dimension spatiale des inégalités environnementales dans les villes est étroitement liée à la répartition spatiale inégale des groupes sociaux dans l’espace. Schaeffer et Tivadar (2019) ont adapté les mesures de ségrégation résidentielle pour évaluer les inégalités environnementales urbaines. Conformément à ce cadre, cet article examine les inégalités basées sur la ségrégation dans les villes françaises, en se concentrant sur l’exposition inégale au couvert végétal (une aménité majeure) et aux sites industriels dangereux (une nuisance majeure). Il révèle si les schémas de ségrégation sont tels que les groupes les plus vulnérables - les travailleurs à faibles revenus, les femmes, les personnes âgées, les familles monoparentales - sont également les plus désavantagés à cet égard.

Haut de page

Texte intégral

Introduction

1The spatial dimension of environmental inequalities in cities is closely linked to the unequal spatial distribution of social groups across neighbourhoods: if social groups were evenly distributed across urban space, their exposure to environmental disamenities and their access to environmental amenities would be much more similar. This means that in cities, distributive environmental justice (EJ), i.e. fairness in the distribution of environmental goods and harms, is linked to socio-spatial segregation, i.e. the spatial separation of social groups.

2Much of EJ research has focused on environmental amenities and inequalities in cities or urban areas. For example, many studies have examined unequal exposure to air pollutants and hazardous waste (e.g. Mah, 2020; Hajat et al., 2015; Carrier et al., 2014; Zwickl et al., 2014; Morello-Frosch and Lopez, 2006; Harner et al., 2002; Chakraborty and Armstrong, 2001; Sheppard et al., 1999). Another set of studies has focused on the uneven distribution of urban green spaces, trees and/or parks (e.g., Neier, 2023; Liotta et al., 2020; Schaeffer and Tivadar, 2019; Byrne, 2017; Apparicio et al., 2016; Frey, 2016; Schwarz et al., 2015; Saporito and Casey, 2015; Shanahan et al., 2014; Zhou and Kim, 2013; Pham et al., 2012; Landry and Chakraborty, 2009). Beyond these examples, EJ research has addressed a wide range of issues, often in urban contexts (see Coolsaet, 2020; Holifield et al., 2017). However, few EJ studies have (i) explicitly considered measures of urban segregation in relation to environmental inequalities, (ii) gone beyond the analysis of a single city or a small set of cities, and (iii) dealt with Western European cities (see Köckler et al., 2017). This article stands at the intersection of these research gaps. In addition, following Bourdieu et al. (1993, cited by Coolsaet and Deldrève, 2024), we consider that ‘the experience of the world’ cannot be reduced solely to socioprofessional (or income) class belonging alone, which remains the dominant lens in French EJ studies. Our analysis considers both class and gender inequalities. Furthermore, to the best of our knowledge, this study is the first to consider environmental inequalities faced by single parents, a group situated at the intersection of gender and income inequalities, since single parents are predominantly women and face poverty far more frequently than other households. Inequality faced by older adults is also taken into account, given the specific vulnerability associated with age.

3In this paper, we measure environmental inequalities in the 97 largest cities of continental France with more than 100,000 inhabitants, focusing on unequal exposure to tree canopy cover (a major amenity) and to dangerous industrial facilities (a major disamenity). Our aim is to test whether patterns of urban segregation are such that the most vulnerable or oppressed groups - low-income workers, women, single-parent families, the elderly - are also the most environmentally disadvantaged.

4In a previous paper, Schaeffer and Tivadar (2019) proposed two segregation-based measures of environmental inequality: the environmental dissimilarity gap index (ΔED) for areal-level environmental data and the environmental centralisation (EC) index for multi-point environmental data. They applied this approach to the agglomeration of Grenoble, France, and provided evidence of the statistically significant unequal distribution of vegetation and industrial risks between poor and non-poor households. Building on this work, we greatly expand the coverage of French cities to allow inter-urban comparisons at a national scale, and we look beyond income-based inequalities to consider the many faces of injustice.

5The next section presents our methodological framework, the empirical background and the data. Section 3 presents and discusses our results. Section 4 concludes.

Empirical Approach

Framework: segregation-based environmental inequalities

6In a review of EJ studies, Mitchell and Walker (2007) found that linear regression was by far the most popular method. More recently, Mennis and Heckert (2017) reviewed the application of spatial statistics, the use of which has been steadily increasing in the field to address the limitations of traditional statistical methods. In contrast, very few EJ studies have used inequality indices to analyse environmental inequalities, with Liotta et al. (2020), Schaeffer and Tivadar (2019), Jacobson et al. (2005) and Lopez (2002) being rare exceptions.

7In particular, Schaeffer and Tivadar (2019) have introduced insights from the residential segregation literature to the EJ field. Conceptually, they refer to segregation-based environmental inequality as a difference between two social groups in terms of their respective degrees of environmental segregation, that is, their geographical separation from an environmental variable. A social group is disadvantaged relative to another group if it is more segregated from an environmental amenity or less segregated from an environmental disadvantage. Methodologically, they proposed two measures based on the dissimilarity index, compatible with areal environmental data, and on the spatial Gini index, compatible with point data (only the location of the amenity is known).

Dissimilarity and Environmental Evenness

8The dissimilarity index (Duncan and Duncan, 1955a) measures the departure from even relative population distribution across spatial units, and it ranges between 0 (evenness distribution of two social groups) and 1 (perfect dissimilarity). It can be interpreted as the share of a group that would have to change its location in order to achieve even relative spatial distribution.

9In the same spirit, the Environmental Dissimilarity index (Schaeffer and Tivadar, 2019) is the dissimilarity between the distribution of a population group and that of an environmental variable across spatial units:

10It can be interpreted as the share of the group that would have to change its location in order to achieve an even spatial distribution with the environmental variable, or in other words, to reduce environmental segregation to zero.

11The segregation-based environmental inequality between social groups is then defined by the differential of environmental dissimilarity:

12A positive value of ΔED means that the group x faces an environmental inequality compared to group y, because it has a more dissimilar distribution with respect to the environmental variable a than group y does.

Gini and Environmental Centralization

13The relative centralisation index (Duncan and Duncan, 1955b) is a specific form of the Gini segregation index and measures the unequal localisation of two groups around a central point, usually the central business district in the context of residential segregation. Tivadar (2019) adapted it for polycentric configurations by taking into account the distance to the nearest point.

14Therefore, Schaeffer and Tivadar (2019) proposed an environmental centralisation index, formally equivalent to the relative centralisation index, but generalised to polycentrism and applied to environmental (dis)amenity:

Data

15To empirically examine the relationship between social segregation and environmental inequalities, we use socio-demographic data from Insee (the French National Statistics Institute) for 97 French agglomeration with more than 100,000 inhabitants. These data, provided at the IRIS 2017 sub-communal level, include 60 social groups (households and population) categorised by socio-professional status, gender, age and marital status, household size and structure, etc.

Table 1. Basic statistics for socio-economic and environmental data at IRIS level

Table 1. Basic statistics for socio-economic and environmental data at IRIS level

16We use two types of environmental data as examples. For areal environmental data, we consider the distribution of tree canopy cover from Copernicus high-resolution data across neighbourhoods. For point data, we use hazardous industrial sites in the metropolitan area, classified as all, hazardous or very hazardous. This geocoded data is provided by the French Ministry of Ecology 2016, as an application of the European Seveso-III Directive (Directive 2012/18/EU) on technological disaster risk reduction. It provides the location of industrial establishments where dangerous substances are used or stored in large quantities, with either a low-risk or a high-risk threshold.

Results

17In this study we focus on segregation and environmental inequalities in relation to two social typologies: vulnerable social groups and low-income households. For the vulnerable social groups, we consider single women and single parent households, and the elderly persons (80 years and over). To measure segregation and environmental inequalities for these categories, we use unigroup indices, which compare the distribution of each group with the rest of the population. For income groups, we use intergroup indices to measure segregation and environmental inequalities between low-income households (i.e. the reference person is a blue-collar worker) and high-income households (i.e. the reference person works in an executive or high intellectual profession).

18In our analysis, we used the OasisR 3.1.1 package (Tivadar, 2024) to compute segregation indices and the SegEnvIneq 1.2 package (Tivadar & Schaeffer, 2024) to calculate environmental inequalities in R software (version 4.4.1).

Segregation patterns in French urban agglomerations

19To analyse the segregation patterns of selected social categories, we use Duncan’s dissimilarity index (Duncan & Duncan, 1955a), but similar results are obtained with the Gini segregation index. After computing the indices in each agglomeration, we plot their distribution in Figure 1.

20As expected, income segregation between workers and managers is high in all metropolitan areas, with an average of 0.33 (on average, 33% of workers households would have to change location to achieve a similar relative distribution to executives). However, there is a high degree of variability, with the index ranging from a minimum of 0.19 to a maximum of 0.51, with a standard deviation of 0.06. The segregation between single women and other households is also high (mean of 0.24) with considerable variability (from 0.09 to 0.32), while the segregation of single parent households is lower (mean of 0.19) and more uniform across agglomerations (standard deviation of 0.03). For the elderly, segregation is quite similar in all agglomerations, with very little variation around the mean of 0.19 (standard deviation of 0.02).

Figure 1. Dissimilarity index distribution in French urban agglomerations.

Figure 1. Dissimilarity index distribution in French urban agglomerations.

21The distribution of the relative centralisation index (Duncan & Duncan, 1955b) is shown in Figure 2. The most centralised category is that of single women, with a mean of 0.13, but with a high dispersion across agglomerations (standard deviation of 0.11). The executive households also have a tendency to be centralized compared to the workers (mean of 0.13), but also with higher heterogeneity, with a standard deviation of 0.12. In the case of single-parent households and the elderly, the preference for central location is less important, with a mean of 0.04 and 0.06 respectively, with a more homogeneous distribution for single-parent households, ranging from -0.06 to 0.15. These patterns could be explained by different factors such as the access to employment, services and social infrastructure, but also by the housing needs. The relative centralisation is significantly positive for all categories (to a lesser extent for the elderly), as confirmed by the Student’s t and Wilcox p-values in Figure 2.

Figure 2. Relative centralization index distribution in French urban agglomerations.

Figure 2. Relative centralization index distribution in French urban agglomerations.

Environmental inequalities analysis

22First, we analyse the environmental inequalities with respect to the highly hazardous industrial sites in France. Since the data are presented as points (their location), we use the environmental centralisation index, the distribution of which is shown in Figure 3. We find that the distribution of environmental centralisation is very similar to that of relative centralisation, except that it is flatter, meaning that the hazardous sites are located relatively centrally in agglomerations. As a consequence, we find that single women and single parents are located closer to these hazardous industrial sites compared to the rest of the households, with significant positive environmental centralisation. In the case of the elderly individuals and executive (vs workers) households, there are no significant environmental inequalities (the p.value of the Student’s t and of Wilcox test are higher than 0.10).

Figure 3. The distribution of environmental centralization in relation to highly hazardous industrial sites.

Figure 3. The distribution of environmental centralization in relation to highly hazardous industrial sites.

23The distribution of the environmental dissimilarity gap relative to tree canopy cover is shown in Figure 4. Again, there are similarities with the relative centralisation index, showing that tree cover increases with distance from the urban centre. The environmental dissimilarity gap is significantly positive for all groups. For single women, single parents and the elderly, the index has positive value in almost all cities (all for single parents): on average, 11% of single women households would have to change location to achieve the same degree of environmental segregation as the rest of the population, 6% of single parents households and 4% of elderly individuals. These groups are therefore clearly disadvantaged in terms of spatial segregation from the tree canopy.

Figure 4. The distribution of environmental dissimilarity gap in relation with tree canopy cover.

Figure 4. The distribution of environmental dissimilarity gap in relation with tree canopy cover.

24On the opposite, workers are less segregated from the tree canopy than executives: on average, 3% of them would have to change location to achieve the same degree of environmental segregation. However, a large proportion of cities display the reverse situation, where workers are more segregated from the tree canopy than executives.

Conclusion

25Using segregation based methodology, this study highlights significant patterns of segregation and environmental inequalities among different social groups in French urban agglomerations.

26The results confirm that income-based segregation remains high, with a clear spatial distinction between workers and executives. Single women also experience notable segregation, with considerable variation across cities, whereas single-parent households and elderly individuals exhibit lower and more uniform segregation levels. The analysis of residential centralization indicates that single women and executives tend to be more concentrated in central areas, likely due to proximity to employment opportunities and urban amenities. Single-parent households and elderly individuals present also a centralisation tendency but they are more evenly distributed, suggesting different housing constraints and preferences.

27Environmental inequality analysis reveals that single women are the most disadvantaged group, both in terms of spatial segregation from the tree canopy and exposure to high-risk hazardous industrial sites. Single parents are also disproportionalely exposed to high-risks and segregated from the tree canopy. The elderly are disproportionalely segregated from the tree canopy, but not overexposed to hazardous sites. Surpringly enough, there are no significant environmental inequalities for executive vs. worker households regarding hazardous sites, and workers are less segregated from the tree canopy than executives in a majority of cities.

28These findings underscore the need for targeted urban policies to reduce segregation and mitigate environmental inequalities, particularly for vulnerable social groups. Further research is required to deepen our understanding of the theoretical and empirical links between residential segregation and related environmental inequalities. To explore the robustness of our results, an extension of this study could involve comparing these findings with additional data obtained at finer and more regular spatial scales, such as using population data from a 200m x 200m grid rather than from the broader French census tracts. Moreover, the vegetation in the uninhabited parts of the study zone also provides valuable ecosystem services to the population and should be considered in future research, to fully capture the environmental benefits across different cities.

Haut de page

Bibliographie

APPARICIO P., PHAM T., SÉGUIN A. & DUBÉ J. (2016), “Spatial distribution of vegetation in and around city blocks on the Island of Montreal: A double environmental inequity, Applied Geography, 76, pp. 128-136.

BOURDIEU P., ACCARDO A., BALAZS G., BEAUD S., BONVIN F., BOURDIEU E. & WACQUANT L. (1993), La misère du monde, Paris, Seuil.

BYRNE J. (2017), “Urban parks, gardens and greenspace, in HOLIFIELD R., CHAKRABORTY J. & WALKER G. (Eds.), The Routledge Handbook of Environmental Justice, London, Routledge, pp. 437-448.

CARRIER M., APPARICIO P., SÉGUIN A. & CROUSE D. (2014), “The application of three methods to measure the statistical association between different social groups and the concentration of air pollutants in Montreal: A case of environmental equity, Transportation Research Part D: Transport and Environment, 30, pp. 38-52.

CHAKRABORTY J. & ARMSTRONG M. P. (2001), “Assessing the Impact of Airborne Toxic Releases on Populations with Special Needs, The Professional Geographer, 53, 1, pp. 119-131.

COOLSAET B. (Ed.) (2020), Environmental Justice: Key Issues, London, Routledge.

COOLSAET B. & DELDREVE V. (2024), “Exploring environmental justice in France: evidence, movements, and ideas, Environmental politics, 33, 5, pp. 757-777.

DUNCAN O. D. & DUNCAN B. (1955a), “A Methodological Analysis of Segregation Indexes, American Sociological Review, 20, 2, pp. 210-219.

DUNCAN O. D. & DUNCAN B. (1955b), “Residential Distribution and Occupational Stratification, American Journal of Sociology, 60, 5, pp. 493-503.

FREY N. (2016), “Equity in the distribution of urban environmental amenities: The case of Washington, D.C., Urban Geography, 38, 10, pp. 1534-1549.

HAJAT A., HSIA C. & O’NEILL M. S. (2015), “Socioeconomic Disparities and Air Pollution Exposure: a Global Review, Current Environmental Health Reports, 2, 4, pp. 440-450.

HARNER J., WARNER K., PIERCE J. & HUBER T. (2002), “Urban Environmental Justice Indices, The Professional Geographer, 54, 3, pp. 318-331.

HOLIFIELD R., CHAKRABORTY J. & WALKER G. (Eds.) (2017), The Routledge Handbook of Environmental Justice, London, Routledge, Taylor & Francis Group.

JACOBSON J. O., HENGARTNER N. W. & LOUIS T. A. (2005), “Inequity Measures for Evaluations of Environmental Justice: A Case Study of Close Proximity to Highways in New York City, Environment and Planning A, 37, 1, pp. 21-43.

KÖCKLER H., DEGUEN S., RANZI A., MELIN A. & WALKER G. (2017), “Environmental justice in Western Europe”, in HOLIFIELD R., CHAKRABORTY J. & WALKER G. (Eds.), The Routledge Handbook of Environmental Justice, London, Routledge, pp. 627-640.

LANDRY S. M. & CHAKRABORTY J. (2009), “Street Trees and Equity: Evaluating the Spatial Distribution of an Urban Amenity”, Environment and Planning A, 41, 11, pp. 2651-2670.

LIOTTA C., KERVINIO Y., LEVREL H. & TARDIEU L. (2020), “Planning for environmental justice: Reducing well-being inequalities through urban greening, Environmental Science and Policy, 112, pp. 47-60.

LOPEZ R. (2002), “Segregation and Black/White Differences in Exposure to Air Toxics in 1990, Environmental Health Perspectives, 110, s2, pp. 289-295.

MAH A. (2020), “Toxic legacies and environmental justice, in COOLSAET B. (Ed.), Environmental Justice: Key Issues, London, Routledge, pp. 121-131.

MASSEY D. S. & DENTON N. A. (1988), “The Dimensions of Residential Segregation, Social Forces, 67, 2, pp. 281-315.

MÉNNIS J. & HECKERT M. (2017), “Application of spatial statistical techniques, in HOLIFIELD R., CHAKRABORTY J. & WALKER G. (Eds.), The Routledge Handbook of Environmental Justice, London, Routledge, pp. 207-221.

MITCHELL G. & WALKER G. (2007), “Methodological issues in the assessment of environmental equity and environmental justice, in Sustainable Urban Development: The Environmental Assessment Methods, vol. 2, London, Routledge, pp. 447-472.

MORELLO-FROSCH R. & LOPEZ R. (2006), “The riskscape and the color line: Examining the role of segregation in environmental health disparities, Environmental Research, 102, 2, pp. 181-196.

NEIER T. (2023), “The green divide: A spatial analysis of segregation-based environmental inequality in Vienna, Ecological Economics, 213, 107949.

PHAM T., APPARICIO P., SÉGUIN A., LANDRY S. & GAGNON M. (2012), “Spatial distribution of vegetation in Montreal: An uneven distribution or environmental inequity?, Landscape and Urban Planning, 107, 3, pp. 214-224.

SAPORITO S. & CASEY D. (2015), “Are There Relationships Among Racial Segregation, Economic Isolation, and Proximity to Green Space?, Human Ecology Review, 21, 2, Canberra, ANU Press.

SCHAEFFER Y. & TIVADAR M. (2019), “Measuring environmental inequalities: Insights from the residential segregation literature, Ecological Economics, 164, p. 106329.

SCHWARZ K., FRAGKIAS M., BOONE C. G., ZHOU W., MCHALE M., GROVE J. M. & CADENASSO M. L. (2015), “Trees Grow on Money: Urban Tree Canopy Cover and Environmental Justice, PLOS ONE, 10, 4, p. e0122051.

SHANAHAN D., LIN B., GASTON K., BUSH R. & FULLER R. (2014), “Socio-economic inequalities in access to nature on public and private lands: A case study from Brisbane, Australia, Landscape and Urban Planning, 130, pp. 14-23.

SHEPPARD E. R., LEITNER H. E., MCMASTER R. B. & TIAN H. O. (1999), “GIS-based measures of environmental equity: Exploring their sensitivity and significance, Journal of Exposure Analysis and Environmental Epidemiology, 9, 1, pp. 18-28.

TIVADAR M. (2019), “OasisR: An R Package to Bring Some Order to the World of Segregation Measurement”, Journal of Statistical Software, 89, 7, pp. 1-39.

TIVADAR M. (2024), OasisR: Outright Tool for the Analysis of Spatial Inequalities and Segregation, R package version 3.1.1, CRAN, https://cran.r-project.org/web/packages/OasisR/index.html

TIVADAR M. & SCHAEFFER Y. (2024), SegEnvIneq: Environmental Inequality Indices Based on Segregation Measures, R package version 1.2, CRAN, https://cran.r-project.org/web/packages/SegEnvIneq/index.html

WEN M., ZHANG X., HARRIS C. D., HOLT J. B. & CROFT J. B. (2013), “Spatial Disparities in the Distribution of Parks and Green Spaces in the USA”, Annals of Behavioral Medicine, 45, S1, pp. 18-27.

ZHOU X. & KIM J. (2013), “Social disparities in tree canopy and park accessibility: A case study of six cities in Illinois using GIS and remote sensing”, Urban Forestry & Urban Greening, 12, 1, pp. 88-97.

ZWICKL K., ASH M. & BOYCE J. K. (2014), “Regional variation in environmental inequality: Industrial air toxics exposure in U.S. cities, Ecological Economics, 107, pp. 494-509.

Haut de page

Table des illustrations

URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-1.png
Fichier image/png, 16k
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-2.jpg
Fichier image/jpeg, 9,6k
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-3.jpg
Fichier image/jpeg, 5,9k
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-4.jpg
Fichier image/jpeg, 12k
Titre Table 1. Basic statistics for socio-economic and environmental data at IRIS level
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-5.jpg
Fichier image/jpeg, 233k
Titre Figure 1. Dissimilarity index distribution in French urban agglomerations.
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-6.jpg
Fichier image/jpeg, 122k
Titre Figure 2. Relative centralization index distribution in French urban agglomerations.
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-7.jpg
Fichier image/jpeg, 157k
Titre Figure 3. The distribution of environmental centralization in relation to highly hazardous industrial sites.
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-8.jpg
Fichier image/jpeg, 159k
Titre Figure 4. The distribution of environmental dissimilarity gap in relation with tree canopy cover.
URL http://journals.openedition.org/belgeo/docannexe/image/82933/img-9.jpg
Fichier image/jpeg, 161k
Haut de page

Pour citer cet article

Référence électronique

Mihai Tivadar et Yves Schaeffer, « Segregation and environmental inequalities in French cities »Belgeo [En ligne], 4 | 2025, mis en ligne le 12 décembre 2025, consulté le 15 janvier 2026. URL : http://journals.openedition.org/belgeo/82933 ; DOI : https://doi.org/10.4000/15c2a

Haut de page

Auteurs

Mihai Tivadar

Corresponding author, Univ. Grenoble Alpes, INRAE, LESSEM, 2 rue de la Papeterie-BP 76, F-38402 St-Martin-d'Hères, France
ORCID 0000-0001-6304-9168
mihai.tivadar@inrae.fr

Yves Schaeffer

Univ. Grenoble Alpes, INRAE, LESSEM, F-38402 St-Martin-d'Hères, France
ORCID 0000-0003-4580-876X
yves.schaeffer@inrae.fr

Haut de page

Droits d’auteur

CC-BY-4.0

Le texte seul est utilisable sous licence CC BY 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont susceptibles d’être soumis à des autorisations d’usage spécifiques.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search