Urban biodiversity and social inequalities in built up cities new evidences, next questions. the example of paris, france

1. Landscape and Urban Planning 106 (2012) 277–287
Contents lists available at SciVerse ScienceDirect
Landscape and Urban Planning
journal homepage: www.elsevier.com/locate/landurbplan
Urban biodiversity and social inequalities in built-up cities: New evidences, next
questions. The example of Paris, France
Marianne Cohena,∗
, Raymond Baudoinb
, Milena Palibrkc
, Nicolas Persynd
, Catherine Rheine
a
Paris Cité-University, Paris Diderot, UMR Ladyss, BP 7001, 75205 Paris Cedex 13, France
b
CBNBP-MNHN, 61 rue Buffon, 75005 Paris, France
c
Paris Cité-University, Paris Diderot, Pôle Image, BP 7001, 75205 Paris Cedex 13, France
d
University Paris 1, Geography Department, 191 rue St Jacques, 75005 Paris, France
e
Paris Cité-University, Paris Diderot, UMR Géographie-cités, BP 7001, 75205 Paris Cedex 13, France
a r t i c l e i n f o
Article history:
Received 4 May 2011
Received in revised form 14 March 2012
Accepted 27 March 2012
Available online 23 April 2012
Keywords:
Urban vegetation
Ecosystem services
Landscape design
Environmental justice
Interdisciplinarity
GIS
a b s t r a c t
The preservation of biodiversity, city-dwellers’ quality of life and equality are major issues in promoting
sustainable cities. We chose to work at the built-up and much-valued heart of an agglomeration: Paris.
We attempt to verify whether public semi-natural spaces with spontaneous vegetation and green frames
provide ecosystem services to city-dwellers in an equitable way, and what is the role played by urban
landscapes in this relation. Based upon existing datasets, a spatial and statistical analysis of relationships
between public semi-natural spaces, urban landscapes and socioeconomic profiles of households was
carried out. We compare their spatial patterns, which vary considerably. Mean income per vegetal and
landscape cluster and correlations between income, ecological parameters and building density show
complex and non-linear relationships. According to Multiple Correspondence Analysis, clusters of public
semi-natural spaces, floristic richness and ecological diversity are associated with specific household
socioeconomic profiles and landscape design. In the western built-up Haussmann districts inhabited by
well-off households, public seminatural spaces provide low ecosystem services. Conversely, in the eastern
and peripheral quarters where lower-income households live, public semi-natural spaces provide higher
ecosystem services. It is not only well-off households which benefit from being close to urban parks
and waterways. Blue and green frames cross districts inhabited by different household profiles. The
implications of such results on inequalities between city-dwellers depends whether the presence of
public semi-natural spaces actually improves their quality of life. These results are highly significant at a
time when urban planners are putting the city’s Biodiversity Plan in place.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
As cities expand in size and spread, a steadily rising num-
ber of people are concerned with the urban way of life. Making
cities fit better with the objectives of sustainable development
is a contemporary challenge (UN-Habitat, 2008). Urban planners
have to promote environmental quality in more compact cities
(EC, 2010; Jabareen, 2006). The presence of public semi-natural
spaces with spontaneous vegetation may provide psychological
benefits (Fuller, Irvine, Devine-Wright, Warren, & Gaston, 2007;
Mayer and McPherson-Frantz, 2004), improvements to human
health (De Vries, Verheij, Groenewegen, & Spreeuwenberg, 2003;
Tzoulas et al., 2007), urban biodiversity and hydro-climatic regula-
tion. In this way, semi-natural spaces support ecosystem services
∗ Corresponding author. Tel.: +33 1 57 27 71 68/71 99; fax: +33 1 57 27 71 74.
E-mail addresses: cohen@univ-paris-diderot.fr (M. Cohen), baudoin@mnhn.fr
(R. Baudoin), rhein@msh-paris.fr (C. Rhein).
and generally, contribute to improving the quality of life in urban
environments (Andersson, 2006; Bolund & Hunhammar, 1999;
Tratalos, Fuller, Warren, Davies, & Gaston, 2007).
However, are these ecosystem services equitably distributed
across the geographic space and equally accessible by all resi-
dents? This question brings up an important concern, given that the
spatial distribution of urban environments is superimposed with
the socio-economic inequalities among the residents (Esponda &
Martinez, 2004; Fainstein, 2010; Pinc¸ on & Pinc¸ on-Charlot, 2004;
Rhein, 1998; UN-Habitat, 2008). Urban environments affect their
residents in many ways, some of which are positive whereas others
are negative, such as pollution. In turn, city-dwellers also modify
the environments in which they live (Pickett et al., 2011). In this
sense, inequalities between city-dwellers may depend on multiple
factors: social, economic, cultural, political, spatial and environ-
mental. According to ‘environmental justice’ studies, minorities
or low-income communities tend to be concentrated in polluted
and distant urban areas (Agyeman, Bullard, & Evans, 2002; Bullard,
2007). Alternatively, other authors suggest that unequal access to
0169-2046/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.landurbplan.2012.03.007

2. 278 M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287
Fig. 1. Location map.
urban biodiversity has implications for the prevalence of ‘envi-
ronmental justice’, therefore representing a challenge for urban
planners (Kinzig, Warren, Martin, Hope, & Katti, 2005; Martin,
Warren, & Kinzig, 2004; Melles, 2005; Strohbach, Haase, & Kabisch,
2009).
Studies conducted in U.S. cities – and, to a lesser extent, in
European cities – reveal a positive correlation between household
incomes and urban biodiversity, which is measured, for instance,
by the number of native and exotic perennial plant species (Kinzig,
2003; Martin et al., 2004), birds (Loss, Ruiz, & Brawn, 2009; Melles,
2005; Strohbach et al., 2009) and other indicators (Kinzig et al.,
2005). Studies have shown that trees are more abundant in dis-
tricts with higher concentrations of wealthy households (Grove &
Burch, 1997; Iverson & Cook, 2000; Talarchek, 1990; Tratalos et al.,
2007). Paradoxically, Barbosa et al. (2007) have found that deprived
and older households living in Sheffield in the U.K. are among the
groups with the greatest access to land parcels that are classified as
natural surfaces (living no further than 300 meters from a natural
surface).
The spatial analysis of urban landscapes may reveal the rela-
tionships between biodiversity and socio-economic profiles, such
as whether wealthier households are located in greener districts
(e.g., Strohbach et al., 2009). Indeed, the number and proportion
of native and exotic species depend on a range of factors related to
urban planning, including the density of buildings and other infras-
tructure (Clergeau, 2007; Kent, Stevens, & Zhang, 1999; Luck, 2007;
McKinney, 2008; Muratet, 2006; Vaquin, Moret, & Le Dantec, 2006;
Williams et al., 2009), the type of “green spaces” (e.g., natural areas
versus parks, Millard, 2008, Strohbach et al., 2009), and the type
of land-use function (e.g., residential versus business or industrial
districts, Dow, 2000; Godefroy & Koedam, 2007; Pickett et al., 2011;
Ricotta, Celesti Grapow, Avena, & Blasi, 2001). Moreover, urban
landscapes are the product of period-specific planning trends that
differ across countries (Conzen, 2004; Harvey, 2005; Loss et al.,
2009; Panerai, Castex, Depaule, & Samuels, 2004; Schwarz, 2010;
Stefulesco, 1993).
While a large body of biodiversity-related research has chosen
a gradient approach, comparing urban, suburban and rural areas,
we chose to study the densely-built core of a conurbation. Previ-
ous research on this area has largely focused on socio-economic
factors, including the negative effects of urban density on human
health (e.g., Murard and Zylberman, 1996), as well as on native veg-
etation (Luck, 2007; McKinney, 2008). However, previous research
has also shown a positive effect of dense urban centers on the
natural environment given their space-saving value and their net
carbon emission output (Jabareen, 2006; Tratalos et al., 2007; EC,
2010).
This paper aims to investigate the spatial distribution of pub-
lic semi-natural spaces and urban landscapes in the urban heart of
a conurbation, as well their socio-economic dimensions, by com-
bining data used in different disciplines (Clergeau, 2007; Grimm
& Redman, 2004; Mathieu, 2009; Pincetl, 2005). First, we charac-
terize the spatial organization and the various features of public
semi-natural spaces in Paris, including the biological traits, the
level of biodiversity, and various functions of ecological services.
Second, we analyze the relationships between urban landscapes,
household socio-economic profiles and public semi-natural spaces.
Findings from this study reveal important implications for urban
planning, particularly in the creation of policies that encompass
conservation issues and equitable access to urban biodiversity for
all households along the socio-economic spectrum (Ahern, 1995;
Hope et al., 2011).

3. M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287 279
2. Materials, methods and context of the study
2.1. Study case
Paris, France is at the heart of one of the densest European
conurbations (Schwarz, 2010; 3596 habitants/km2 based on cal-
culations by INSEE Census in 2008). The City of Paris is a relatively
small geographic area (105 km2) and has a fairly constant popula-
tion size (2 211 297 inhabitants), in comparison to the conurbation
sprawls that have grown in both geographic area (Paris urban unit
2008: 2865 km2; 1999–2008: +4.4%) and population size (popu-
lation 2008: 10 303 382; 1999–2008: +6.8%, Fig. 1). Paradoxically,
the high urban density in Paris has not precluded its high standard
of living (e.g., high property prices: Beckouche & Roudier, 1992;
Morlet, 2000; Touati, 2010) or even the extent of its biodiversity
(2900 vegetal and animal species, Vaquin et al., 2006). Similarly to
many cities worldwide, urban planners and policymakers in Paris
have to reconcile issues related to biodiversity and social partici-
pation (Biodiversity Plan, Green and Blue frame Plan) with more
general objectives of urban planning (i.e., Grand Paris Plan).
2.2. Overview of the method
To explore the distribution patterns of urban landscapes, the
socio-economic profiles of households and the availability of public
semi-natural spaces, we used GIS (Geographical Information Sys-
tem) in addition to statistical analysis. The study area was divided
into 992 Census-Block Groups (referred to as CBG). CBG is the basic
spatial container of information (i.e., spatial unit) used by INSEE to
store census data. This spatial unit layer was used as a common
spatial grid to perform our analysis.
Fig. 2 presents an overview of the analytical framework. In the
first step, we explored each dataset by performing a clustering
analysis on urban landscapes, household profiles and public semi-
natural spaces data to differentiate between sub-groups sharing
similar properties (Sections 2.3–2.5). Second, we assessed the var-
ious ecosystem services provided to individuals and communities
through the availability of public semi-natural spaces, based on
their functions for ecosystem maintenance, hydro-climatic regula-
tion and socio-cultural externalities (Bolund & Hunhammar, 1999,
§ 2.6). Third, we explored the links between urban landscapes,
household socio-economic profiles and public semi-natural spaces.
The results are presented in Section 3 and discussed in Section 4.
2.3. Urban landscapes data
2.3.1. APUR database
High resolution urban landscapes data (1 m) were provided by
the APUR (Paris Urban Planning Agency). The layers are derived
from infra-red aerial ortho-photography and Lidar data analysis
(performed by Inter Atlas Corporation) and geographical databases.
These data describe the spatial distribution of the following geo-
graphic features: (1) vegetation patches per strata (i.e., <1 m,
1–10 m, >10 m); (2) water bodies, bare soil and asphalt; and (3)
built up areas based on the median height of buildings and the
period of construction (i.e., 11 periods dating from ‘pre-1850’ to the
present).
2.3.2. Intersecting landscape patches and Census-Block-Groups
We intersected this layer with the CBG layer and computed the
following variables per CBG: the percentage of ground-cover by
built areas, water, soil, asphalt and vegetation; the distribution of
built areas per periods and the distribution of vegetation coverage
per strata; and the mean and maximum heights of buildings.
2.3.3. Urban landscape clusters
An analysis based on hierarchical clustering (Euclidian distance,
Ward method, using Xl’stat software) identifies 10 types of urban
landscapes (Appendix A), even though the ‘mean Parisian land-
scape’ was densely built during the nineteenth century. The cluster
analysis differentiates between the old ‘mineral’ center where the
pre-industrial bourgeoisie lived, the dense Haussmann districts
that were built for the bourgeoisie of the second Industrial Revolu-
tion (during the nineteenth century) and the dense inner suburbs
that were inhabited by workers and artisans (Harvey, 2005; Garden
& Pinol, 2009; Jordan, 1995). The social housing and facilities belt
was built beginning in the 1920s; some modern dwellings were
built even later according to the Athenian Chart (Panerai et al.,
2004). Natural elements and tree cover – concentrated in parks, in
woods and along waterways – tend to decrease over time in built
areas.
2.4. Household data
2.4.1. INSEE Census database
The INSEE Census of 2008 provides data per CBG: the number of
households (approximately 1000), the surface area, the distribution
of 8 professions and social categories, and the mean fiscal revenue
by consumption unit (henceforth referred to as the mean income).
2.4.2. Household socio-economic profiles
Hierarchical clustering (Euclidian distance, Ward method,
Xl’stat software), applied to the distribution of socio-professional
categories, identified seven profiles; the eighth category corre-
sponds to uninhabited CBGs and is therefore excluded from the
calculations (Appendix B). In terms of the socio-professional pro-
file of Parisian households, the overwhelming majority comprises
heads of household in managerial and professional occupations
(31%) and retirees (22%); however, the gap between mean incomes
computed per cluster is marked on a scale of 1–5. Low and unskilled
blue- and white-collar workers are largely represented (50%) in the
lowest income cluster (no. 7, D 14 271). By comparison, heads of
household in managerial and professional occupations, business
leaders, trade people and inactive people represent 51% of house-
holds in the highest income cluster (no. 6, D 71 976). The income
ratio is 178 between the mean incomes of the lowest tenth of
the poorest CBG (D 1423) and the highest tenth of the richest CBG
(D 253 209). This revenue structure explains the gap between mean
(D 35 210) and median (D 26 146) income and suggests a hierar-
chy in the household profiles and, more generally, the existence
of social inequalities, as observed by Rhein (1998), Esponda and
Martinez (2004) and Pinc¸ on and Pinc¸ on-Charlot (2004).
2.5. Public semi-natural space data
2.5.1. FLORA database
We extracted the following data from the FLORA database
(Vaquin et al., 2006): botanical surveys operated in public areas
between 1992 and 2010, polygons containing more than 3 species
(n = 774) and vegetal species with more than two occurrences
(n = 673). This selection strategy aims to eliminate obsolete surveys
and random data. Surveyed vegetal species consist of native species,
sub-spontaneous species, or naturalized species and planted trees.
2.5.2. Public semi-natural space clusters
K-Means Clustering (Package stats, R software version 2.9.2) (R
Development Core Team, 2009) identified 6 types of public semi-
natural spaces according to their botanical composition, biological
traits (Julve, 1998) and floral richness: entomophilous wastelands
(median 90 taxa), woods, hygrophilous waterways banks, cemeter-
ies, gardens (35–50) and sidewalks (10) (Decaudin, Cohen, Baudoin,

4. 280 M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287
Fig. 2. General approach, methods and findings.
& Palibrk, 2012). Despite its density, the city of Paris provides a wide
range of habitats for flora, comparable to those observed in western
outer suburbs (Muratet, 2006).
2.5.3. Ecosystem services of public semi-natural spaces
To evaluate the societal functions of public semi-natural spaces,
we used the concept of ecosystem services. We considered three
specific functions of ecosystems for the benefit of individual and
communal wellbeing, in accordance with Bolund and Hunhammar
(1999), and assessed these functions using several sources and ref-
erences (Appendix C). The maintenance function of ecosystems
was assessed by their biological traits (specifically, inferring biotic
interactions and adaptation to water); their hydro-climatic regula-
tion function was assessed by the proportion of engineer-species
(Phanerogames) and sealing soil; and their cultural function was
assessed by the use and perception of vegetation clusters by city-
dwellers. The combination of these functions enables us to qualify,
rather than quantify, the ecosystem services provided by each clus-
ter from low to excellent.
2.5.4. Intersecting the FLORA polygons with the
Census-Block-Groups
The intersection between the FLORA polygons and the CBG layer
selected 282 CBGs with botanical information. The CBGs with no
botanical information may be due to polygons containing less than
3 species (which were omitted from the analysis), a lack of a survey
or the absence of spontaneous vegetation. When two or more types
of semi-natural space were present in a CBG, we applied a major-
ity filter. An ecological diversity index was calculated for each CBG,
according to the number and spatial distribution of vegetation clus-
ters (using the Shannon formula). Floral richness was calculated by
spatial aggregation at the CBG level to adjust for differences in sam-
ple size and the diversity of habitats in the Flora database inventory,
which is highly correlated with floral richness. We further classified
these two values in 5 classes (Appendix D).
2.6. Cross-relating of thematic data
To explore the relationship between urban landscapes, house-
hold profiles and public semi-natural spaces, we performed a
statistical analysis on spatially explicit variables that measure both
quantitative and qualitative dimensions. Our analytical procedure
is as follows:
• We described and compared the spatial patterns of each thematic
cluster.
• On the basis of the botanical information available for 282
CBGs, we computed the mean household income per botanical
and landscape cluster, in addition to assessing the correlation
between mean revenue, floral richness, the ecological diversity
index and building density.
• We conducted a multiple correspondence analysis, which can
ascertain the relationships between the thematic clusters,
according to the values of the significance test of the modality
variables on the factorial axes. We selected the following vari-
ables for the analysis: urban landscape types (10 modalities),
household profiles (8 modalities), public semi-natural spaces (6
modalities), class of floral richness (5 modalities) and class of
ecological diversity (5 modalities).
• We compared the rank of the hierarchy of ecosystem services
and that of household profiles per CBG. Then, we mapped these
configurations and calculated the proportions of households and
surface area in each configuration.
• We used the GIS buffer function to identify the 186 CBGs located
around uninhabited CBGs (e.g., woods, parks and waterways). We
perform the Pearson’s Chi-squared Test in creating the contin-
gency tables to verify the relationship between socio-economic
profiles of households and vicinity to uninhabited CBGs.
3. Results
3.1. Comparing spatial patterns of urban landscapes, household
profiles and public semi-natural spaces
The spatial patterns of our three datasets are different as they are
influenced by distinct historical circumstances (Fig. 3). This further
indicates the complex and paradoxical relationships between the
three sets (Sections 3.2–3.4).
3.1.1. Spatial pattern of urban landscapes
The urban landscape of Paris is imperfectly organized based
on a center-periphery model that followed different trends of
urban planning and decreasing building density (Fig. 3a) as the
city expanded over time beyond successive walls and fortifica-
tions. Superimposed on this concentric pattern, there is a distinct
west-east division, as Haussmann districts were built westwards,
in an effort to avoid industrial pollution by facing westerly winds
(Garden & Pinol, 2009). Peripheral arrondissements, which are
typically densely built, combine inner suburbs, green districts,
Haussmann avenues and modern dwellings. They contrast with the
less dense belt of public housings and facilities, built upon the site
of the last fortification. Urban parks, gardens and cemeteries, which
are scattered all over the city or located in its vicinity (e.g., woods)

5. M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287 281
Fig. 3. Spatial pattern – a: landscape – b: socio-economic profiles – c: semi-natural spaces.
as well as running alongside the banks of the Seine River and the
canals, are sparsely located.
3.1.2. Spatial pattern of household profiles
The wealthiest households are concentrated in the west side
of the city (e.g., cluster 6 is mostly composed of high-income resi-
dents in managerial and professional occupations, business leaders,
traders, retired and inactive people). However, moving outwards
along concentric rings, the socio-economic profiles of the residents
become increasingly heterogeneous. These concentric rings are
bounded by a belt with indentations along inner railways, inhab-
ited by low- or unskilled blue- and white-collar workers, with
lower incomes (clusters 4 and 7, Fig. 3b). This pattern of spatial
organization is largely due to historical circumstances, notably the
displacement of the bourgeoisie from the old center to the west dur-
ing the nineteenth century (Garden & Pinol, 2009; Harvey, 2005).
The wealthiest households still live in these western districts,
whereas lower-income households reside in the social-housing
and facilities belt (Pinc¸ on & Pinc¸ on-Charlot, 2004). Conversely, the
inner suburbs can be characterized by five different household pro-
files, with high income variations (D 19 206–41 066). Esponda and
Martinez (2004) argue that this income heterogeneity is a feature
of the Parisian social structure.
3.1.3. Spatial pattern of public semi-natural spaces features
According to our multi-criteria assessment, the ecosystem ser-
vices of public semi-natural spaces are higher in the peripheral
woods, parks, wastelands and cemeteries, medium along water-
ways and lower in the central gardens and sidewalks. There is a
similar gradient for floral richness and ecological diversity (Fig. 3c).
Historical circumstances partly explain this spatial pattern, as
observed in Rome (Ricotta et al., 2001) and Brussels (Godefroy &
Koedam, 2007). Woods, which were royal hunting grounds until
the French Revolution, are located on the edges of the city, whereas
French-style gardens and esplanades are concentrated in the old
city center. Haussmanian Parks, railways and cemeteries were
moved to or built in the city’s ‘periphery’ during the nineteenth
century. Since the 1970s, urban renovation and disuse of infrastruc-
ture have generated wastelands (e.g., along the deserted railway,
Petite Ceinture, which constitutes a green frame). Non-operating
industrial plants in the city’s peripheries have been converted into
parks (Fig. 1). Meanwhile, the declining use of chemical herbicides
by municipal technical services over the last decade has facilitated
the growth of opportunistic species in certain sections of public gar-
dens and along sidewalks. Ecological factors play a lesser role, as
observed in other cities (Pickett et al., 2011); even along waterways,
the hygrophilous vegetation grows within a built framework and
is periodically removed by waterway technical services to protect
against building damage.
3.2. Relationships among household incomes, urban landscapes
and public semi-natural spaces
Further analysis based on the quantitative data show complex
and non-linear relationships among household incomes, the bio-
diversity of public semi-natural spaces and building density; yet,
we generally anticipated these results given that their respective
spatial patterns were quite different.
3.2.1. Mean household incomes per type of public semi-natural
space
Mean household incomes decrease by nearly half (1.8) accord-
ing to the function and design of public semi-natural spaces.
Household incomes are higher among CBGs where woodland
(D 52 343) or waterside vegetation (D 49 073) are widely prevalent
and, conversely, lower in CBGs dominated by unplanned public
semi-natural spaces (cemeteries: D 30 194; wasteland: D 29 337).
In CBGs where garden and sidewalk vegetation are prevalent,
incomes approximate the citywide mean (respectively D 37 486
and D 35 092). Mean incomes vary slightly, but with no particular
order, across classes of ecological diversity (1.2) and floral richness
(1.4).

6. 282 M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287
3.2.2. Mean household income per urban landscape cluster
The mean household income decreases by a ratio of 2.4 across
urban landscape clusters, according to urban planning trends;
levels of building density and the historical periods of build-
ing establishment are less clearly related to household income.
The increasing density of the two Haussmann types (standard:
41%, dense: 53%) is inversely associated with mean income
(D 49 166–38 045) and tree-coverage (9–2%). Modern dwellings
(37%) are more dense than green districts (31%), despite similar
mean values of household income among residents of both areas
(around D 26 000). Despite the similar levels of building density
(45%) and periods of establishment (nineteenth century), house-
holds in inner suburbs have a lower mean income (D 30 737) than
those in Haussmann districts and the old center (D 38 045–49 166,
density 41–55%). Mean incomes are lowest in the outer belt
(D 19 009–22 593), which are also the less densely built areas (21%).
3.2.3. Relationships among floral richness, ecological diversity,
household income and building density
The relationships among mean income, floral richness
(r2 = 0.007) and ecological diversity (r2 = 0.08) are non-linear
and not statistically significant (L-shaped scatter plot, Fig. 4).
Ecological parameters vary widely across socio-economic group-
ings (less so for low-income profiles: nos. 4 and 7), along with a
wide income range of wealth profiles (nos. 3, 5 and 6). With the
exception of uninhabited CBGs (class 8), the mean values of floral
richness and ecological diversity are close to the centroid of the
scatter plot for socio-economic classes.
Similarly, all of the relationships among building density and flo-
ral richness (r2 = 0.292), ecological diversity (r2 = 0.136) and mean
incomes (r2 = 0.02) are non-linear.
3.3. Relationships among urban landscapes, household profiles
and public semi-natural clusters
The first four axes of the multiple correspondence analysis
extract 77% of the total variance and display complex organiza-
tional factors. Plan 1–2 shows a Guttman effect (horseshoe effect),
which highlights the opposition of the average values (axis 2)
to the extreme values (axis 1, Fig. 5). The first axis represents a
gradient of floral richness and ecological diversity and an opposi-
tion between adapted (hygrophilous vegetation) and opportunistic
vegetation (on sidewalk). This ecological gradient is associated
with a socio-demographic hierarchy (i.e., uninhabited CBG group
versus inhabited CBGs with wealthy households, composed of
those mainly working in managerial and professional occupations,
with a mean income of D 41 066) and a building density gradient
(1% for Seine banks versus 41–55% for the dense districts built for
the bourgeoisie during the nineteenth century or even before in the
old center). On the mid-gradient, axis 2 highlights wastelands with
high ecosystem services in districts, which were originally designed
for workers and small craftsmen during the nineteenth century
but are currently mainly inhabited by working- and middle-class
households (mean incomes of D 14 271–24 088). Here, the build-
ing density is moderately high (20–45%). The third axis contrasts
wastelands and sidewalks with moderate ecological services of gar-
den vegetation, associated with densely built Haussmann districts
(41–53%, built during the nineteenth century) and wealthy house-
holds (mean income of D 41 066). Woods and cemeteries, which are
poorly represented, are highlighted in axes 4 and 5 (3.6 and 2.8% of
variance).
3.4. Ecosystem services and socio-economic inequalities
There is a difference between the rank of the hierarchy of
ecosystem services and that of household profiles for 73% of
households and 76% of the surface area of surveyed CBGs (Fig. 6).
Wealthy households (mean incomes of D 38 799–71 976) that are
located in the central and western districts of Paris are associated
with ecosystem services evaluated as low or medium (sidewalk
and garden vegetation; 37% of households, 31% of the surface
area). Similarly, middle-class household profiles (mean incomes
of D 24 088–29 653) tend to be located in areas with low ecosys-
tem services (sidewalk vegetation, 14% of households, 11% of
the surface area). Conversely, middle- and working-class house-
holds, which are located in the eastern parts of Paris and in
the outer belt (mean incomes of D 14 271–29 653), are associ-
ated with high ecosystem services (22% of households, 34% of
the surface area). The Petite Ceinture, a deserted railway line, has
an important ecological function –as a corridor for insects-, and
has a broad social appeal that benefits the residents of the outer
arrondissements, who come from wide-ranging socio-economic
backgrounds.
3.5. Proximity of uninhabited public woods, parks and waterways
CBGs containing waterways, parks and cemeteries are uninhab-
ited. Their ecosystem services are associated with two different
household profiles located within their proximity, as confirmed
by a highly significant Chi-squared Test result (p-value < 0.0001,
df = 7, ˛ = 0.05, Chi-square: critical value: 14.07, observed value:
33.76). The wealthiest households (mean income of D 71 971, con-
tribution to Chi-square: 31%) are located near French-style gardens,
esplanades and the riverbanks of the Seine in the western parts of
Paris, whereas middle-income households tend to reside around
parks and along the waterside in the eastern part (mean income
of D 29 653, contribution to Chi-square: 29%). Waterways have
an important ecological function – water facilitates pollination
and the dispersion of hygrophilous plants – and an important
social function, as they provide many households, regardless
of their socio-economic profiles, with breath-taking views of a
‘blue vista’.
4. Discussion
In the following discussion, we address two key points: first, the
relationships among public semi-natural spaces, socio-economic
inequalities and urban landscapes; and second, the implications of
our findings that point to new avenues of research.
4.1. Urban biodiversity and social inequalities, new evidence in
the built-up city
There are distinct associations between household socio-
economic profiles and the types of public semi-natural spaces,
although ecosystem services are more important in districts
inhabited by middle and working-class households than in those
inhabited by wealthy households. Our study confirms the relation-
ships – which are well established in the urban ecology literature
– between vegetation, urban landscapes (Dow, 2000; Luck, 2007;
McKinney, 2008) and socio-economic factors (Pickett et al., 2011).
However, we do not find a linear positive relationship between
household income and species richness, as other studies have
(Kinzig, 2003; Kinzig et al., 2005; Loss et al., 2009; Martin et al.,
2004; Melles, 2005; Strohbach et al., 2009). It is not only high-
income households, but also middle-class households, who live in
close proximity to public parks and waterways; this result is close
to those obtained by Barbosa et al. (2007). The relationship between
the abundance of trees and the level of household income, which
has been established in other studies (Grove & Burch, 1997; Iverson
& Cook, 2000; Talarchek, 1990; Tratalos et al., 2007), is only evident
in two types of Haussmann districts.

7. M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287 283
Fig. 4. Relationships among household income, number of species (A), ecological diversity (Shannon Index) (B) and socio-economic clusters (framed numbers 1–8). Low
values of ecological parameters are associated with a wide range of incomes and lower incomes with a wide range of ecological parameters, so there is no linear relationship
between these values.
The differences between the findings of our study and past
research are partly linked to methodological divergences. If all
exotic species are included in the species count or if the sam-
ple includes private gardens, maintenance costs would explain
why those plants are mostly found in wealthy areas. In these
cases, income may be a key factor, as reported in other studies
(Kinzig, 2003; Kinzig et al., 2005; Loss et al., 2009; Martin et al.,
2004; Strohbach et al., 2009) and, more generally, in environmen-
tal justice studies (Helfand & Peyton, 1999). If one considers only
native species, the linear relationship between species richness
and household income is negative (native birds in Chicago, Loss
et al., 2009). However, in our case, the species count does not
include all exotic species (i.e., shrubs, bushes, annual and perennial
herbs are excluded), and consequently, we find no significant
linear relationship. If private spaces and all exotic plants were
included in botanical inventories, then our results might have been
different.
Another source of the differences between our results and those
of previous studies is linked to different urban planning preroga-
tives across time, which resulted in substantial variations in the
socio-economic profiles of households distributed across urban
landscapes. Given that wealthier households live in “greener” res-
idential suburbs (such as in the USA and Canada: Kinzig, 2003;
Kinzig et al., 2005; Loss et al., 2009; Martin et al., 2004; Melles,
2005) or in districts with a large amount of urban green coverage
(in Leipzig: Strohbach et al., 2009; in Brussels: Godefroy & Koedam,
Fig. 5. First factorial plan of multiple correspondence analysis.

8. 284 M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287
Fig. 6. Ecosystem services and socio-economic clusters.
2007), it is unsurprising that they are associated with higher biolog-
ical diversity (McKinney, 2008). These factors appear to be inversely
correlated in a city, such as Paris, that is marked by the ‘Haussmann
paradox’, i.e., the fact that the upper social categories live in high
density districts with low vegetation coverage (Palibrk & Rhein,
2011; Touati, 2010).
Urban planning trends seem to have more influence on eco-
logical parameters than building age (Loss et al., 2009) or density
(Clergeau, 2007; Kent et al., 1999; Luck, 2007; McKinney, 2008;
Muratet, 2006; Williams et al., 2009), except in the case of uninhab-
ited parks, woods and waterways. Our results are more aligned with
those found in studies by Godefroy and Koedam (2007), Millard
(2008) and Snep, Van Ierland, and Opdam (2009) about the function
of landscapes and ‘green spaces’ in biodiversity. Our results are sim-
ilar to the findings of Pellissier, Cohen, Boulay, and Clergeau (2012)
about the role of urban landscape configurations on the abundance
of bird guilds in Paris, as these configurations are linked to urban
planning trends.
The Haussmann architectural and vegetation model (Stefulesco,
1993) designed for the Parisian bourgeoisie is unfavorable for
the presence of public semi-natural spaces with high levels
of biodiversity and ecosystem services, despite the declin-
ing use of herbicides in public spaces. The model turned the
city into the “Capital of Modernity” during the nineteenth
century (Harvey, 2005), but this model has since lost some
comparative advantage. Conversely, wasteland and cemeteries
are located in peripheral non-Haussmanian districts where the
working- and middle-classes live. This spatially and socially
oriented model recognizes that ecosystem services are more
important in districts where middle and working-class house-
holds live than in relatively wealthier districts. Waterways and
non-operating railways contribute to this phenomenon when
they cross districts with more varied socio-economic household
profiles.
4.2. New questions arising from our findings
We hesitate to draw strong conclusions from our results because
a statistical association between variables does not ascertain
causality or interactions between them. Indeed, ecosystem services
provided by public semi-natural spaces are more important in areas
inhabited by middle and working-class households than in rela-
tively wealthy areas. Deserted railways and waterways, designing
ecological corridors, are attractive features to households living in
close proximity, regardless of their socio-economic profiles. How-
ever, what are the implications of such opposite or transverse
spatial patterns on inequalities between city-dwellers? It remains
merely speculation that the presence of urban biodiversity actually
improves quality of life, bringing psychological and health benefits
as suggested by De Vries et al. (2003), Fuller et al. (2007), Mayer
and McPherson-Frantz (2004) and Tzoulas et al. (2007). The actual
influence of environmental factors on inequalities between city-
dwellers requires further research. Our approach does not account
for the complexity of interactions between city-dwellers and their
natural environments. We considered the use and perception of
public semi-natural spaces in the assessment of ecosystem ser-
vices from an all-encompassing perspective that is not spatially
explicit. Not all residents regard semi-natural spaces positively
(Boutefeu, 2005; Nassauer, 1995). The views of residents depend
on what is meant by ‘urban nature’, their lifestyle behaviors and
socio-economic profiles (Breuste, 2004; Kinzig et al., 2005) and the
type of district where they live (Snep et al., 2009).
Some wealthy Parisians who live in Haussmann districts do
not develop a strong sensory relationship with the city. Instead,
their need for ‘nature’ is fulfilled by going to their country houses
or to private facilities of the Bois de Boulogne, and their ‘well-
being’ is addressed by the interior facilities of their residences
(Grésillon, 2010). It is also questionable if the ‘green vista’ pro-
vided by the Petite Ceinture is appreciated by all types of households.

9. M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287 285
In the wealthy sixteenth arrondissement, this disused railway con-
verted in a ‘Nature Path’ is not used much, as residents use other
public garden close by (Fig. 1). The landscape along the Seine
and the canals is popular among visitors, but the hygrophilous
vegetation growing close to the water is mostly inconspic-
uous. These considerations moderate the significance of our
findings.
However, the call for more contact with urban nature and the
political response to that class does appear to be widespread in the
non-Haussmanian districts, based on the distribution of voluntary-
run shared gardens (Fig. 1). Another exemplar of this is the wild
seed campaign (‘laissons pousser’) that was launched in 2010 in
partnership with Naturparif, the Biodiversity Agency and the local
authority, for which the center of activity is mainly located in the
eastern parts of the city (Fig. 1). According to our survey con-
ducted in the eastern twentieth arrondissement, 62% of respondents
usually go to urban sites with abundant spontaneous flora, 21%
prefer conventional parks or outer woods and 17% do not go to
any “green spaces” regularly. Thus, high ecosystem services of
public semi-natural spaces are most likely appreciated by middle
and working-class households living in these districts and con-
tribute to improving their quality of life, which strengthens our
findings.
Another limitation of our study arises from the datasets. Our
objectives called for more thorough floristic and other species-
driven datasets, as well as more detailed socio-economic and
census data. Socio-economic data were not sufficiently precise (8
instead of 41 socioprofessional categories, as in the previous INSEE
Census of 1999) to assess households’ socio-cultural profiles and
the way that they may appreciate urban nature, as other studies
have analyzed (Barbosa et al., 2007; Kinzig et al., 2005; Palibrk &
Rhein, 2011; Snep et al., 2009). Research on a world-class city, such
as Paris, calls for a large-scale study to determine the influence of its
social and spatial stratification on the use and perception of urban
nature. Moreover, data for vegetation in private spaces and cer-
tain types of exotic species were unavailable. A final limitation of
our study is that areas with species counts of less than 3 and non-
surveyed areas were excluded. Improving the spatial sampling of
vegetation of the FLORA database and the inclusion of exotic species
would certainly strengthen our analysis.
5. Conclusion and perspectives for urban planning
Should public semi-natural spaces and green frames serve as
democratizing elements for city-dwellers, whilst the gap between
the mean incomes of households grouping suggests that Paris is
an ‘unjust’ city (Fainstein, 2010)? Our results are mixed. On the
one hand, our findings suggest that ecosystem services of public
semi-natural spaces should benefit middle- and working-class
households more than wealthier households, mainly because of
landscape design and function rather than building density and
age. On the other hand, our results are inconclusive regarding
whether the presence of public semi-natural spaces and green
frame actually improves the residents’ quality of life, which
depends on how residents enjoy and even increase the presence of
that vegetation, depending on their perception of nature, lifestyle
behaviors and socio-economic profile. Some city-dwellers consider
‘nature’ as part of their urban environment, such as those who
supported the Biodiversity Plan (Mairie de Paris, 2010), participate
in voluntary-run shared garden and wild-seed campaigns (Fig. 1),
care for individual micro-gardens (Blanc, Cohen, & Glatron, 2007)
or visit public semi-natural spaces. Nevertheless, according to
our inquiries, the visual appearance of wasteland might not be
enjoyed by certain residents (because of its poor aesthetic quality,
dirtiness or lack of planning). Moreover, wealthy households have
alternative ways of compensating for the lack of ‘nature’ in the
districts where they live. Therefore, public semi-natural spaces and
green frames may be more beneficial for less wealthy city-dwellers
who do not possess a second home in the countryside.
The strategy for how to address public semi-natural spaces, such
as wastelands, is rather inconsistent. Some are not open to the
public. Others are preserved to protect certain species and trans-
formed into natural or voluntary-run shared gardens by residents
and/or planners (Fig. 1). In addition, there are certain public
semi-natural spaces that have disappeared altogether under new
buildings. These issues – related to biodiversity, social partici-
pation, quality of life, and environmental justice – are deemed
irrelevant when urbanization projects are underway. For exam-
ple, the wasteland on the rue de Fontarabie in the twentieth
arrondissement, formerly a playground for the neighborhood chil-
dren, has disappeared to make way for a smaller conventional
garden surrounded by new buildings associated with a crèche.
Other wastelands such as the Petite Ceinture (a non-operating rail-
way) are preserved because of unresolved issues of land ownership.
Although two sections have been transformed into a footpath and
others are used unofficially, the majority of the network remains
as is, which makes it possible to preserve the green frame in the
form of a rare ecological corridor for insect life in Paris. It also has a
social function, as its green vista is shared by many nearby house-
holds across the income gradient. To a lesser extent, waterways
play a similar role.
During the implementation of the City of Paris’s Biodiversity
Plan, we presented some of our results at the ‘Nature in Paris’
Workshop, organized by the Paris urban planning agency (APUR,
2011) with the involvement of the local Biodiversity Department
(DEVE). Quality of life, environmental justice and democracy con-
stitute important motivations to raise certain groups’ awareness,
such as those who do not take part in elaborating the Biodiver-
sity Plan, and to reinforce the political legitimacy of protecting
biodiversity. Communicating with city-dwellers and educating
children are also important issues; public semi-natural spaces
could find social ‘utility’ as places of learning (e.g., educational
‘insect’ or ‘butterfly spots’ in entomophilous wastelands). Uses
such as these are considered priorities by city-dwellers based on
survey research and feedback from public consultations for the Bio-
diversity Plan (Mairie de Paris, 2010). Paradoxically, our results
suggest that urban planners and city-dwellers should get more
involved in questions of biodiversity – which may be paradoxi-
cal in that the development of public semi-natural spaces is largely
spontaneous.
The field of urban ecology studies, then, remains open, and
more research is needed to reconcile ecological function with city-
dwellers’ quality of life and sense of equality.
Acknowledgements
This paper is part of MEEDEM-CNRS (Program PIRVE), and
National Research Agency (ANR, Trames Vertes Urbaines Pro-
gram). We would like to thank Etienne Gresillon, Julien Lefour,
Laurent Simon and Lydie Goeldner (inquiries), Michel Godron,
master students (Arnaud Samba-Fouani, Yann Brunet and Barbara
Decaudin), Chantal Rémon (computation). Edward Hughes trans-
lated our manuscript to English and anonymous reviewers helped
improve the paper. APUR Parisian Urban-planning Agency and
Inter-Atlas Corporation (agreement with UMR Ladyss-University
Paris Diderot), INSEE (Statistical Department, agreement with Cen-
tre Maurice Hallbachs-CNRS), DGI (Fiscal Agency, agreement with
INSEE) provide databases, Pôle Image (University Paris Diderot) the
technical support.

10. 286 M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287
Appendix A. Average profile of Paris landscape and of the
10 landscape clusters
Paris and clusters Building in % Asphalt in % Height of buildings
in m.
Main period of
construction
Vegetation + water + soil in % Trees in %
Mean Max
Paris mean landscape 41 30 10.8 34.1 1851–1914 29 6
Old center 55 32 11.6 28.6 Before 1850 13 3
Dense Haussmanian 53 34 11.2 29.1 1851–1914 13 2
Haussmanian 41 31 11.7 31.7 28 9
Inner suburbs 45 28 12.2 33.5 27 5
Modern dwellings 37 29 11.5 70.8 1960–1970 34 6
Green districts 31 28 9 35.9 Miscellaneous 41 6
Social housing belt 21 38 2.8 23.9 1915–1939 41 9
Facilities belt 20 36 7.3 37.5 1915–1950 44 7
Parks, woods 11 11 6.4 28.6 – 78 19
Banks 1 17 1 6.5 – 82 8
Appendix B. Average profile of Paris households and of the
8 socio-professional clusters
Clusters Paris mean
1 2 3 4 5 6 7 8
Mean revenue in D 24 088 29 653 41 066 19 206 38 799 71 976 14 271 35 210
Households professions in %
Business leaders, trade people, craftsmen 3 3 5 4 4 8 3 Less than 60
households
4
Managerial and professional occupations 24 33 38 21 35 32 9 31
Intermediate occupations 17 20 14 19 14 9 14 15
Unskilled white-collars 15 13 10 18 9 10 22 12
Blue collar workers 8 7 4 13 3 3 14 6
Retired people 26 17 18 15 27 26 30 22
Non-working people 7 7 11 11 7 11 7 9
Appendix C. Multi-criteria valuation of ecosystem services
of semi-natural spaces, from low to excellent
Type of services Criteria Types of semi-natural spaces
Wood Sidewalk Bank Garden Wasteland Cemetery
Self-maintenance Biotic interactions, adaptation to watera
+++ + +++ + +++ ++
Cultural Uses and perceptionsb
+++ + ++ +++ ++ ++
Regulation Engineer-species,c
sealing soild
+++ + ++ +
Synthetic index Excellent Low Medium High
a
Deduced from the significant biological traits (Decaudin et al., 2012).
b
Inferred from inquiries.
c
Assessed according to the significant proportion of Phanerophytes, considered
as engineer-species by Grime (1998); according to field survey.
Appendix D. Clusters of floral richness and ecological
diversity
CLAS1 CLAS2 CLAS3 CLAS4 CLAS5
Floral richness 3–16 17–46 47–74 75–118 119–373
Ecological diversity 0 0.001–0.23 0.24–0.47 0.48–0.69 0.7–1.24
References
Agyeman, J., Bullard, R. D., & Evans, B. (2002). Exploring the nexus: Bringing together
sustainability. Environmental Justice and Equity. Space and Polity, 6(1), 77–90.
Ahern, J. (1995). Greenways as a planning strategy. Landscape and Urban Planning,
33, 131–155.
Andersson, E. (2006). Urban landscapes and sustainable cities. Ecological Society,
11(1), 34.
APUR (2011). Situation et perspectives de la place de la nature à Paris (The place
of Nature in Paris: Position and prospect). URL: http://www.apur.org/etude/
situation-perspectives-de-place-de-nature-paris (in French).
Barbosa, O., Tratalos, J. A., Armsworth, P. R., Davies, R. G., Fuller, R. A., Johnson, P., et al.
(2007). Who benefits from access to green space? A case study from Sheffield,
UK. Landscape and Urban Planning, 83, 187–195.
Beckouche, P., & Roudier, J. (1992). Les revenus fiscaux des ménages, un indicateur
des disparités territoriales [Household fiscal incomes, an indicator of spatial
differences]. Mappemonde, 2(92), 18–25 (in French)
Blanc, N., Cohen, M., & Glatron, S. (2007). What role does plant landscape play in
urban policy? In M. Berlan-Darqué, Y. Luginbuhl, & D. Terrasson (Eds.), From
landscape knowledge to landscape management (pp. 83–99). Paris: QUAE Cema-
gref.
Bolund, P., & Hunhammar, S. (1999). Ecosystem services in urban areas. Ecological
Economics, 29, 293–301.
Boutefeu, E. (2005). La demande sociale de nature en ville, enquête auprès des habi-
tants de l’agglomération lyonnaise. In Coll. Recherche 154 Condé-sur-Noireau (in
French).
Breuste, J. H. (2004). Decision making, planning and design for the conservation of
indigenous vegetation within urban development. Landscape and Urban Plan-
ning, 68, 439–452.
Bullard, R. D. (Ed.). (2007). Growing smarter. Achieving livable communities, environ-
mental justice, and regional equity. Cambridge: MIT Press.
Clergeau, Ph. (2007). (Urban landscape ecology) Une écologie du paysage urbain. Paris:
Apogée. (in French).

11. M. Cohen et al. / Landscape and Urban Planning 106 (2012) 277–287 287
Conzen, M. R. G. (Ed.). (2004). Thinking about urban form. Papers on urban morphology,
1932 –1998. Bern: Peterlang.
Decaudin, B., Cohen, M., Baudoin, R., Palibrk, M. (2012). Analyse des relations entre
données géographiques et données floristiques en vue d’une cartographie de
la Trame Verte et Bleue dans la Ville de Paris [Analysis of the relationship
between geographical and botanical data for the mapping of green frames in
Paris City]. Seminar Approches méthodologiques en géomatique pour la car-
tographie de la Trame Verte et Bleue, GDR MAGIS, 24 janvier 2012 – Paris. URL:
http://magis.ecole-navale.fr/ (in French).
De Vries, S., Verheij, R. A., Groenewegen, P. P., & Spreeuwenberg, P. (2003). Natural
environments – healthy environments? Environmental Planning, 35, 1717–1731.
Dow, K. (2000). Social dimensions of gradients in urban ecosystems. Urban Ecosys-
tems, 4, 255–275.
Esponda, M., & Martinez, C. (2004). A Paris, les ménages les plus aisés voisins des
plus modestes. In Insee-Apur, Insee Ile-de-France à la. (p. 24) (in French)
European Commission. (2010). World and European Sustainable Cities, Insights from
EU research, Directorate-General for Research Socio-economic Sciences and Human-
ities. EUR 24353 EN.
Fainstein, S. S. (2010). The just city. New York: Cornwell University Press.
Fuller, R. A., Irvine, K. N., Devine-Wright, P., Warren, P. H., & Gaston, K. J. (2007).
Psychological benefits of green space increase with biodiversity. Biology Letters,
3, 390–394.
Garden, M., & Pinol, J. L. (2009). (Parisians Atlas, from the French Revolution to nowa-
days) Atlas des Parisiens de la Révolution à nos jours. Paris: Parigramme. (in
French).
Godefroy, S., & Koedam, N. (2007). Urban plant species patterns are highly
driven by density and function of built-up areas. Landscape Ecology, 22,
1227–1239.
Grésillon, L. (2010). (Feeling the town) Sentir la ville. Paris, QUAE: Coll. Indisciplines.
(in French).
Grime, J. P. (1998). Benefits of plant diversity to ecosystems: Immediate, filter and
founder effects. Journal of Ecology, 86, 902–910.
Grimm, N. B., & Redman, C. L. (2004). Approaches to the study of urban ecosystems:
The case of Central Arizona—Phoenix. Urban Ecosystems, 7, 199–213.
Grove, J. M., & Burch, W. R., Jr. (1997). A social ecology approach and applications of
urban ecosystem and landscape analyses: A case study of Baltimore, Maryland.
Urban Ecosystems, 1, 259–275.
Harvey, D. (2005). Paris capital of modernity. New York: Routledge.
Helfand, G. E., & Peyton, L. J. (1999). A conceptual model of environmental justice.
Social Science Quarterly, 80(1), 68–83.
Hope, D., Gries, C., Zhu, W., Fagan, W. F., Redman, C. L., Grimm, N. B., et al. (2011). Plan-
ning and design of ecological networks in urban areas. Landscape and Ecological
Engineering, 7(1), 27–32.
Iverson, L. R., & Cook, E. A. (2000). Urban forest cover of the Chicago region
and its relation to household density and income. Urban Ecosystems, 4,
105–124.
Jabareen, Y. R. (2006). Sustainable urban forms: Their typologies, models, and con-
cepts. Journal of Planning Education and Research, 26(1), 38–52.
Jordan, D. P. (1995). Transforming Paris: The life and labors of Baron Haussmann.
Chicago: University of Chicago Press.
Julve, Ph. (1998) ff. – Baseflor. Index botanique, écologique et chorologique de
la flore de France (Botanical, ecological and chorological index of French
Flora). Version: “date de la version citée”. URL: http://perso.wanadoo.fr/
philippe.julve/catminat.htm (in French).
Kent, M., Stevens, R. A., & Zhang, L. (1999). Urban plant ecology patterns and pro-
cesses: A case study of the flora of the city of Plymouth, Devon, UK. Journal of
Biogeography, 26, 1281–1298.
Kinzig, A. (2003). Socioeconomics drives urban plant diversity. Proceedings of
the National Academy of Sciences of the United States of America, 100(15),
8788–8792.
Kinzig, A. P., Warren, P., Martin, Ch., Hope, D., & Katti, M. (2005). The effects of
human socioeconomic status and cultural characteristics on urban patterns of
biodiversity. Ecological Society, 10(1), 23.
Loss, S. R., Ruiz, M. O., & Brawn, J. D. (2009). Relationships between avian diver-
sity, neighborhood age, income, and environmental characteristics of an urban
landscape. Biological Conservation, 142, 2578–2585.
Luck, G. W. (2007). A review of the relationships between human population density
and biodiversity. Biological Reviews, 82, 607–645.
Mairie de Paris. (2010). (Biodiversity white book in Paris) Livre blanc de la biodiversité
à Paris. Paris: Mairie de Paris. (in French)
Martin, C., Warren, P., & Kinzig, A. (2004). Neighbourhood socio-economic status is
a useful predictor of perennial landscape vegetation in residential neighbour-
hoods and embedded small parks of Phoenix AZ. Landscape and Urban Planning,
69, 355–359.
Mathieu, N. (2009). Constructing an interdisciplinary concept of sustainable
urban milieu, 2009 Compass Interdisciplinary Virtual Conference. URL:
http://compassconference.wordpress.com/2009/10/29/conference-paper-
constructing-an-interdisciplinary-concept-of-sustainable-urban-milieu/
(March 2011).
Mayer, F. S., & McPherson-Frantz, I. (2004). The connectedness to nature scale: A
measure of individual’s feeling in community with nature. Journal of Environ-
mental Psychology, 24, 503–515.
McKinney, M. L. (2008). Effects of urbanization on species richness: A review of
plants and animals. Urban Ecosystems, 11, 161–176.
Melles, S. J. (2005). Urban bird diversity as an indicator of human social diversity
and economic inequality in Vancouver, British Columbia. Urban Habitats, 3(1),
25–48.
Millard, A. (2008). Semi-natural vegetation and its relationship to designated urban
green space at the landscape scale in Leeds, UK. Landscape Ecology, 23(10),
1231–1241.
Morlet, O. (2000). Marché du logement et ségrégation spatiale en région parisienne
[Housing market and social segregation in Paris region]. Etudes Foncières, 85, 8–9
(in French)
Murard, L., & Zylberman, P. (1996). (Health in the Republic) Hygiène dans la République.
Paris: Fayard. (in French).
Muratet, A. (2006). Diversité végétale en milieu urbain: l’exemple des Hauts-de-Seine
(Urban vegetal diversity: The example of Hauts de Seine). Thèse de Doctorat Univer-
sité. Paris VI, Paris: CBNBP-MNHN. (in French).
Nassauer, J. I. (1995). Messy ecosystems, orderly frames. Landscape Journal, 14,
161–170.
Palibrk, M., & Rhein, C. (2011). Formes urbaines, structures sociales et végétation à
Paris. Des relations contre-intuitives (Urban forms, social structures and vegetation
in Paris. Some unexpected relationships), Oral communication, Colloque Images et
Villes. University Paris Diderot-Campus Spatial, 12–13 December (ongoing paper
in Cybergeo).
Panerai, P., Castex, J., Depaule, J.-C., & Samuels, I. (2004). Urban forms: Death and life
of the urban block. Oxford: Architectural Press.
Pellissier, V., Cohen, M., Boulay, A., & Clergeau, Ph. (2012). Birds are sensitive to
landscape composition and configuration also within the city centre. Landscape
and Urban Planning, 104, 181–188.
Pickett, S. T. A., Cadenasso, M. L., Grove, J. M., Boone, C. G., Groffman, P. M., Irwin, E.,
et al. (2011). Urban ecological systems: Scientific foundations and a decade of
progress. Journal of Environmental Management, 92, 331–362.
Pincetl, S. (2005). La durabilité urbaine et la nature en ville: le besoin
d’interdisciplinarité. In N. Mathieu, & Y. Guermond (Eds.), La ville durable, du
politique au scientifique (pp. 209–220). Paris: Quae Editions (in French).
Pinc¸ on, M., & Pinc¸ on-Charlot, M. (2004). (Sociology of Paris) Sociologie de Paris. Paris:
La Découverte. [in French].
R Development Core Team (2009). R: A language and environment for statisti-
cal computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN
3-900051-07-0. URL: http://www.R-project.org.
Rhein, C. (1998). Globalization, social change and minorities in Metropolitan Paris:
The emergence of new class pattern. Urban Studies, 35(3), 429–447.
Ricotta, C., Celesti Grapow, L., Avena, G., & Blasi, C. (2001). Topological analysis of
the spatial distribution of plant species richness across the city of Rome (Italy)
with the echelon approach. Landscape and Urban Planning, 57, 69–76.
Schwarz, N. (2010). Urban form revisited—Selecting indicators for characterizing
European cities. Landscape and Urban Planning, 96(1), 29–47.
Snep, R., Van Ierland, E., & Opdam, P. (2009). Enhancing biodiversity at business sites:
What are the options, and which of these do stakeholders prefer? Landscape and
Urban Planning, 91(1), 26–35.
Stefulesco, C. (1993). (The green urbanism) L’urbanisme végétal. Paris, Mission du
Paysage: IDF. (in French)
Strohbach, M. W., Haase, D., & Kabisch, N. (2009). Birds and the city: Urban biodi-
versity, land use, and socioeconomics. Ecological Society, 14(2), 31.
Talarchek, G. M. (1990). The urban forest of New Orleans—an exploratory analysis
of relationships. Urban Geography, 11, 65–86.
Touati, A. (2010). Histoire des discours politiques sur la densité [History of political
discourse upon density]. Etudes Foncières, 145, 24–26 (in French).
Tratalos, J., Fuller, A. R., Warren, P. H., Davies, R. D., & Gaston, K. J. (2007). Urban form,
biodiversity potential and ecosystem services. Landscape and Urban Planning,
83(4), 307–318.
Tzoulas, K., Korpela, K., Venn, S., Yli-Pelkonen, V., Ka´zmierczak, A., Niemela, J.,
et al. (2007). Promoting ecosystem and human health in urban areas using
Green Infrastructure: A literature review. Landscape and Urban Planning, 81(3),
167–178, 20.
UN-Habitat. (2008). State of the World’s Cities 2008/2009. Harmonious Cities. London:
Earthscan.
Vaquin, J. B., Moret, J., & Le Dantec, J. P. (2006). (Nature Atlas in Paris) Atlas de la nature
à Paris. Paris: Le passage APUR. (in French).
Williams, N. S. G., Schwartz, M. W., Vesk, P. A., McCarthy, M. A., Hahs, A. K., Clemants,
S. E., et al. (2009). A conceptual framework for predicting the effects of urban
environments on floras. Journal of Ecology, 97, 4–9.
Cohen Marianne, Lecturer and doctoral thesis supervisor, Paris Cité-University,
Paris Diderot, within the laboratory UMR 7533 Ladyss. Director of PPF Pôle Image.
Specialized in the domains of biogeography, relationships between vegetation, nat-
ural resources and societies, GIS, in urban and rural areas.
Baudoin Raymond, Teacher and researcher to the National natural history museum
of Paris (MNHN) within the national botanical Conservatoire of Paris Basin, unit of
the department Ecology and Biodiversity Management. Specialized in the domains
of the computing, databases and statistics.
Palibrk Milena, Engineer, Paris Cité-University, Paris Diderot within the PPF
Pôle Image. Specialized in the domains of GIS, remote-sensing and geographical
databases.
Persyn Nicolas, PhD student, University Paris 1, within the laboratory UMR 8504
Géographie-cités, CRIA. Specialized in urban planning, urban sustainability and
land-use management.
Rhein Catherine, Research Director, Paris Cité-University, Paris Diderot, within the
laboratory UMR 8504 Geographie-cités. Specialized in the domains of social geog-
raphy and statistics.