Environ. Res. Lett. 17 (2022) 104035 https://doi.org/10.1088/1748-9326/ac9325 LETTER Exploring green gentrification in 28 global North cities: the role OPEN ACCESS of urban parks and other types of greenspaces RECEIVED 6 July 2022 Margarita Triguero-Mas1,2,3,∗, Isabelle Anguelovski2,3,4, James J T Connolly2,3,5, Nick Martin2,3, REVISED Austin Matheney2,3, Helen V S Cole2,3, Carmen Pérez-Del-Pulgar2,3,6,7, Melissa García-Lamarca2,3, 12 August 2022 Galia Shokry2,3,8, Lucía Argüelles3,9, David Conesa10, Elsa Gallez2,3,11, Blanca Sarzo12,13, ACCEPTED FOR PUBLICATION 20 September 2022 Miguel Angel Beltrán14, Jesúa López Máñez15,16, Joaquín Martínez-Minaya17, Emilia Oscilowicz2,3, PUBLISHED Mariana C Arcaya1 and Francesc Baró2,3,11 4 October 2022 1 Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, MA, United States of America 2 Institute for Environmental Science and Technology, Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Spain 3 Original content from Barcelona Lab for Urban Environmental Justice and Sustainability, Barcelona, Spain this work may be used 4 Institucio Catalana de Recerca i Estudis Avancats, Barcelona, Spain under the terms of the 5 School of Community and Regional Planning, University of British Colombia, Vancouver, Canada Creative Commons 6 Attribution 4.0 licence. Department Environmental Policy, Helmholtz Centre for Environmental Research (UFZ), Leipzig, Germany 7 Department of Political Science, Friedrich-Schiller-University Jena, Fürstengraben 1, 07743 Jena, Germany Any further distribution 8 of this work must School of Environmental and Sustainability Science, Kean University, NJ, United States of America 9 maintain attribution to Universitat Oberta de Catalunya (UOC), Estudis d’Economia i Empresa and Internet Interdisciplinary Institute (IN3), Barcelona, the author(s) and the title Spain of the work, journal 10 Valencia Bayesian Research Group, University of Valencia, Valencia, Spain citation and DOI. 11 Vrije Universiteit Brussel (VUB), Geography and Sociology Departments, Brussels, Belgium 12 Department of Microbiology and Ecology, University of Valencia, Valencia, Spain 13 School of Mathematics and Maxwell Institute for Mathematical Sciences, University of Edinburgh, Edinburgh, United Kingdom 14 Fundación para el Fomento de la Investigación Sanitaria y Biomédica de la Comunitat Valenciana, Valencia, Spain 15 Departament of Financial Economics, University of Valencia, Valencia, Spain 16 CaixaBank Group, Barcelona, Spain 17 Department of Applied Statistics and Operational Research, and Quality, Universitat Politècnica de València, Valencia, Spain ∗ Author to whom any correspondence should be addressed. E-mail: mtrigueromas@gmail.com Keywords: greenspace, gentrification, green infrastructure, nature-based solutions, environmental justice Supplementary material for this article is available online Abstract Although cities globally are increasingly mobilizing re-naturing projects to address diverse urban socio-environmental and health challenges, there is mounting evidence that these interventions may also be linked to the phenomenon known as green gentrification. However, to date the empirical evidence on the relationship between greenspaces and gentrification regarding associations with different greenspace types remains scarce. This study focused on 28 mid-sized cities in North America and Western Europe. We assessed improved access to different types of greenspace (i.e. total area of parks, gardens, nature preserves, recreational areas or greenways [i] added before the 2000s or [ii] added before the 2010s) and gentrification processes (including [i] gentrification for the 2000s; [ii] gentrification for the 2010s; [iii] gentrification throughout the decades of the 2000s and 2010s) in each small geographical unit of each city. To estimate the associations, we developed a Bayesian hierarchical spatial model for each city and gentrification time period (i.e. a maximum of three models per city). More than half of our models showed that parks—together with other factors such as proximity to the city center—are positively associated with gentrification processes, particularly in the US context, except in historically Black disinvested postindustrial cities with lots of vacant land. We also find than in half of our models newly designated nature preserves are negatively associated with gentrification processes, particularly when considering gentrification throughout the 2000s and the 2010s and in the US. Meanwhile, for new gardens, recreational spaces and greenways, our research shows mixed results (some positive, © 2022 The Author(s). Published by IOP Publishing Ltd Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al some negative and some no effect associations). Considering the environmental and health benefits of urban re-naturing projects, cities should keep investing in improving park access while simultaneously implementing anti-displacement and inclusive green policies. 1. Introduction and also in East Asia (e.g. Cheonggyecheon Stream Restoration Project or the Gyeongui Line Forest Park Fostered by global sustainability agendas (e.g. UN in South Korea) (Loughran 2014, Sandberg 2014, SDG 11), cities globally are increasingly mobiliz- Haase et al 2017, Kwon et al 2017, Weber et al 2017, ing re-naturing projects towards green, sustainable, Immergluck and Balan 2018, Jo Black and Richards and healthy urban environments (Haase et al 2017). 2020, Anguelovski and Connolly 2021). Those projects create or restore existing greenspaces To implement re-naturing agendas and pro- and other vegetated environments, including parks, jects able to address socio-environmental and health greenways or gardens, as well as blue spaces like challenges while preventing gentrification and dis- waterfronts (Haase et al 2017, Taylor and Hochuli placement, we must first understand how prevalent 2017, de Oliveira and Mell 2019). The adoption of re- green gentrification is, the conditions under which it naturing projects in cities is often motivated by exist- occurs, and the magnitude of its effects (Pearsall and ing evidence that green and blue spaces can address Eller 2020). Yet, empirical quantitative evidence on diverse urban socio-environmental and health chal- the relationship between greenspaces and gentrifica- lenges, including urban heat island effects, air pol- tion remains scarce, as much of the literature to date lution, water regulation and increased prevalence has been dominated by qualitative case study ana- of non-communicable diseases (Kabisch et al 2017, lyses of a specific city and/or of a particular neighbor- de Oliveira and Mell 2019, Veerkamp et al 2021, Yang hood (Curran and Hamilton 2012, Loughran 2014, et al 2021). Anguelovski and Connolly 2021). Within this gen- At the same time, the implementation of re- eral paucity of quantitative studies investigating the naturing interventions may contribute to the attrac- relationships between greenspaces and gentrification tion of new investment and place these areas at higher at the citywide and cross-city levels, the few stud- risk for gentrification—defined as socio-cultural and ies that do exist show mixed findings (Irwin 2002, physical exclusion and displacement of underpriv- Conway et al 2010, Kwon et al 2017, Anguelovski et al ileged residents (i.e. minority racial/ethnic groups, 2018, Maantay and Maroko 2018, Connolly 2019, Du working class individuals, non-white residents and and Zhang 2020, Pearsall and Eller 2020, Rigolon and immigrants) from the neighborhood (Glass 1964, Németh 2020, Shokry et al 2020, Chen et al 2021, Brown-Saracino 2010)—through a process known as Donovan et al 2021, Schinasi et al 2021, Kim and green or environmental gentrification (Dooling 2009, Wu 2022, Anguelovski et al 2022). Existing quantit- Gould and Lewis 2017). The mechanisms by which ative research commonly utilize either single socio- re-naturing projects induce or accelerate gentrifica- economic attributes to assess gentrification or one tion are material and discursive. Greening contrib- socio-economic variable at a time (Conway et al 2010, utes to increased property and land values as well Kwon et al 2017, Du and Zhang 2020, Pearsall and as urban green grabbing practices (i.e. developers’ Eller 2020, Kim and Wu 2022), the most common of appropriation of the financial and social benefits gen- these socio-economic variables being property values erated by nearby new or planned green amenities) (Conway et al 2010, Du and Zhang 2020). (Safransky 2014, Czembrowski and Kronenberg 2016, Further, few studies consider the heterogeneity García-Lamarca et al 2022). Re-naturing projects are of urban greenspace size, vegetation cover, species further articulated around a green rhetoric that serves richness, amenities, legal protection, and other socio- to (re-)brand cities (i.e. urban green boosterism) environmental characteristics. A growing body of lit- and eventually produce commodified environmental erature suggests that large-scale urban greening pro- amenities (Garcia-Lamarca et al 2019, Triguero-Mas jects such as the BeltLine in Atlanta or the 606 et al 2021). These projects may thus produce new rails-to-trails project in Chicago may induce envir- types of social injustices (Anguelovski 2016, Rigolon onmental gentrification as they may be planned—at and Németh 2018). least partly—to increase property values (Loughran Green gentrification has been clearly documented 2014, Kwon et al 2017, Immergluck and Balan 2018, in a diversity of urban areas, particularly linked with Jo Black and Richards 2020, Chen et al 2021). How- flagship interventions, such as projects in the US (e.g. ever, a study of ten large and medium-size US cit- the High Line Park in New York City or the BeltLine in ies found that park size did not have an impact on Atlanta), in Europe (e.g. the Lene-Voigt Park in Ger- gentrification (Rigolon and Németh 2020). Mean- many, the Noorderpark in Amsterdam or the Lim- while, smaller greenspaces, are conceptualized as part hamn Quarry National Natural Reserve in Sweden) of a ‘just green enough’ approach that may not 2 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al trigger gentrification (Chen et al 2021) but may these effects varied over time, between mid- and long- secure environmental and public health benefits of re- term gentrification, and between cities representing a naturing interventions (Wolch et al 2014). variety of geographies and urban development con- More specifically, different greenspace charac- texts. Through our analysis, we derive city-level find- teristics are related to greenspace types (e.g. parks, ings from tract-level observations. allotments or biodiversity areas) (Smith et al 2017) which may in turn lead to different gentrification pat- 2. Methods terns due to their diverse uses and variety of res- idents’ appreciation (Amorim Maia et al 2020). For 2.1. Study design example, parks may be perceived as spaces of daily We designed a multi-city study to assess improved use for a variety of activities and may be strongly access to different types of greenspace and gentri- linked to neighborhood identity. Meanwhile, allot- fication processes over 1990–2016 in mid-sized cit- ment or community gardens and biodiversity-rich ies in North America and Western Europe (supple- areas such as nature preserves may be used as selling mental material, page 3). With these criteria, we points, marketed as community spaces or valued for selected those mid-sized cities with strong greening their environmental and biological worth. Less than agendas since the 1990s and available and reliable a handful of studies (most of them from the US) data on the variables of interest for our study, while have considered the associations between typologies aiming to include cities that represented different of green spaces and property prices (Lutzenhiser and regions and different growth trajectories. This pro- Netusil 2001, Czembrowski and Kronenberg 2016). cess led to including 28 cities in this study: 17 in However, scholars have clearly stated the importance North America (including 14 cities in the US: Atlanta of understanding how greenspace typologies may be (GA), Austin (TX), Baltimore (MD), Boston (MA), linked to green gentrification (Pearsall and Eller 2020, Cleveland (OH), Denver (CO), Detroit (MI), Louis- Stuhlmacher et al 2022). ville (KY), Milwaukee (WI), Philadelphia (PA), Port- There is only scarce evidence on whether green land (OR), San Francisco (CA), Seattle (WA), Wash- gentrification is solely or more strongly linked with ington DC (DC); and also three cities in Canada: certain greenspace types and most of this evidence Calgary, Montréal, Vancouver), and 11 in Western focuses in the US. The study by Rigolon & Németh Europe (Lyon and Nantes in France; Bristol, Shef- reported that greenways located within half a mile field and Edinburgh in the United Kingdom; Ams- of census tract centroids increased the odds of the terdam in The Netherlands; Barcelona and Valencia census tract gentrification by over 200% (Rigolon and in Spain; Copenhagen in Denmark, Dublin in Ireland Németh 2020). A study in Portland showed that each and Vienna in Austria) (figure 1). new street tree planted in a census tract was associated with a $265 increase in tract-level median sales price 2.2. Assessment of gentrification 12 years after tree planting, thus impacting neighbor- In the absence of one universally accepted quantitat- hood affordability (Donovan et al 2021). Similarly, ive measure of gentrification, researchers have used existing community gardens have been associated a range of methods to assess if places are gentrifying with increased gentrification in Brooklyn, New York and how rapidly (Freeman 2005, Ding et al 2016). (Maantay and Maroko 2018) and in St.Louis, Mis- In this study we calculated a gentrification score souri (Braswell 2018). According to another recent per small geographical unit (SGU) in each city and study from New York City, passive (not associated time period using a unified methodology described with exerting recreational activities), natural (non- previously (Anguelovski et al 2022). In brief, our manicured, wild) greenspaces contribute to gentrific- gentrification score is based on several measures of ation processes (Kim and Wu 2022). In the specific sociodemographic change and change in rent val- case of parks, one study in the north-eastern half of ues. To build our gentrification score we calculate Barcelona found that areas around new urban parks diversity in change across several sociodemographic experienced gentrification, but not in neighborhoods constructs (ethnic/racial vulnerability, higher educa- with housing units difficult to upgrade (Anguelovski tion, income, poverty, professional occupation) plus et al 2018). change in a real estate indicator (rental housing) by Considering this mixed evidence on green gentri- using Shannon’s Equitability Index. That is, accord- fication, we hypothesized that each type of greenspace ing to our composite score, gentrification occurs at might play a different role in green gentrification the highest degree when several indicators change in processes. Consequently, this study aimed to explore a strong direction toward gentrification at the same the association between different types of greenspaces time (Anguelovski et al 2022). and gentrification using quantitative data from a total Accordingly, we obtained the most closely aligned of 28 cities in the Global North. More specifically, variables that corresponded to the sociodemographic we investigated whether different types of greenspaces and real estate constructs previously stated for 2000, (parks, gardens, nature preserves, recreational spaces, 2010 and 2016 (or the closest date possible to these and greenways) predict gentrification and whether years) in each SGU and city from national and local 3 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Figure 1. Situation map of the cities included in this study. statistics offices (see tables S1–S3). Based on these preserves, recreational areas or greenways added to data, we obtained two indicators of mid-term gentri- each city SGU during the time period before the fication (gentrification for the 2000s, gentrification 2000s (i.e. before the start of the variable ‘gentri- for the 2010s) and one indicator for long-term gentri- fication in the 2000s’) plus 2 years (table S4). We fication (gentrification throughout the decades of the included the 2 years overlap to account for the impact 2000s and 2010s) (supplemental material, page 19). that announcing a greenspace creation or improve- ment may have on gentrification (Immergluck 2009). 2.3. Assessment of new greenspaces of different The last five indicators (Parks pre ∼ 2010s, Gar- typologies dens pre ∼ 2010s, Preserves pre ∼ 2010s, Recreational We obtained data for each city on different types of pre ∼ 2010s, Greenways pre ∼ 2010s) indicate the same new greenspaces opened for public use between 1990 as the previous five categories, but for the time period and 2016 using a combination of direct communica- before the 2010s and 2 years during the time-period tions with city and non-profit organizations, city land included in that gentrification variable, following the records, published reports, media reports, historic same logic as for the pre ∼ 2000s indicators. imagery and city archival comprehensive searches (Connolly and Anguelovski 2021, Anguelovski et al 2.4. Covariates 2022). We considered that new greenspaces were Based on previous literature, we selected our study all those substantially vegetated publicly accessible covariates. Distance from the SGU centroid to the spaces that had been inaugurated or publicly acquired city centre and SGU population density per square (i.e. greenspaces that were formerly private spaces kilometer in 2010 and change in population dens- and were transferred to public ownership) during our ity between the different time points we studied was period of study (i.e. 1990–2016). calculated by the research team using ArcGIS 10.6 Our study included five greenspace categories software. SGU green coverage prior to 1990 was also based on generic cross-city exclusionary definitions estimated using existing databases and ArcGIS 10.6 that built on common typologies of urban green- software. We also assessed the number of public transit space (Cvejic et al 2015, Smith et al 2017): parks, stops added to the SGU immediately before and dur- gardens, nature preserves, recreational areas, and ing the first 2 years of the studied decade using Google greenways (find a summary of the classification in maps, Wikipedia and city level planning documents table 1, for a detailed description see (Connolly and to identify the location of all new rail and bus stops. Anguelovski 2021, Anguelovski et al 2022) and sup- The number of new residential buildings built in the plemental material, page 20). tract during each time period analyzed was derived Based on identified data we built the following ten from national and state statistics office reports, local indicators: Parks pre ∼ 2000s, Gardens pre ∼ 2000s, building permit databases and lot-level tax or cadas- Preserves pre ∼ 2000s, Recreational pre ∼ 2000s, tral files. Greenways pre ∼ 2000s, Parks pre ∼ 2010s, Gar- dens pre ∼ 2010s, Preserves pre ∼ 2010s, Recre- ational pre ∼ 2010s, Greenways pre ∼ 2010s. The 2.5. Statistical analyses first five indicators (Parks pre ∼ 2000s, Gardens We developed Bayesian hierarchical spatial models for pre ∼ 2000s, Preserves pre ∼ 2000s, Recreational estimating the relationships between gentrification pre ∼ 2000s and Greenways pre ∼ 2000s) denote the and our variables of interest (i.e. new greenspaces), total area of—respectively—parks, gardens, nature while adjusting them for the relevant covariates. More 4 Table 1. Typology of greenspaces considered in this study, including definition, sub-types and relevant aspects when evaluating green gentrification. New greenspace type Definition Spaces comprised in this type Relevant aspects when evaluating green gentrification Parks • Accessible, high-quality opportunities for informal • Comprises all urban parks, ornamental • They offer a broad range of health promotion and recreation and community events. These spaces gardens, parterres, and green plazas recreational possibilities, including enhancing social Environ. Res. Lett. 17 (2022) 104035 include a variety of features such as formal footpaths, that are designated as vegetated areas. networks with neighbors. These spaces are also a play space, sports areas, trees, planted beds and/or prominent space of urban biodiversity (Cvejic et al ponds. 2015). • We considered that these spaces may increase neighborhood attractiveness as parks may be perceived as spaces to use daily to promote and 5 prevent one’s health. Gardens • Spaces reserved for a defined set of gardeners to grow • Includes community and allotment • These greenspaces are spaces for social interactions, food or aromatic plants. gardens. while being places of food provisioning and physical activity (Cvejic et al 2015). • We considered that these spaces may be attractive to the first waves of gentrifiers as spaces to socialize and feel part of the neighborhood community. Nature preserves • Areas designated as protected from development and • Includes urban forests, biodiversity • These spaces are biodiversity hotspots, at the same allowed to grow naturally in a relatively undisturbed areas, conservation areas, nature time as they provide cooling, air pollution, carbon state. These areas are residual natural habitats, planted reserves and protected areas. sequestration and water regulation. They may also be or vacant land that has been colonized by vegetation important source of certain foods and raw materials and wildlife. They are often meant to protect (Cvejic et al 2015). biodiversity. • We considered that these spaces may increase • In our study, when there was a doubt about the legal neighborhood attractiveness for those particularly protection status of the greenspace, the space was interested in and those who value wildlife considered a park instead of a nature preserve. conservation, biodiversity and environmental education or those interested in living in less densely populated areas, surrounded by larger greenery. (Continued.) M Triguero-Mas et al Table 1. (Continued.) Environ. Res. Lett. 17 (2022) 104035 New greenspace type Definition Spaces comprised in this type Relevant aspects when evaluating green gentrification Recreational • Natural and artificial playing surfaces. Large and • Sport fields opened for use (e.g. not • These spaces are particularly relevant as physical spaces generally flat areas of grassland or specially designed within an enclosed stadium) and activity promoting amenities (Smith et al 2017). surfaces, used primarily for designated sports municipal recreation centers largely • We considered that these spaces may not prominently 6 i.e. playing fields, tennis courts, bowling greens. comprised of attached green areas. lead to green gentrification, as other similar grey spaces may provide similar services to residents. Greenways • Linear greenspaces designed for people to move • Includes riparian, natural and other • These spaces may have very different impacts through them. They may link different areas within a vegetated corridors, greenbelts, and according to their vegetation. Certain greenways city as part of a designated and managed network and linear promenades with substantial structures may increase bird and insect richness and used for walking or cycling, or linking urban vegetation. abundance or provide cooling, wind-control, air greenspaces and/or or cities to their surrounding pollution and carbon sequestration (Cvejic et al cities, towns, countryside or country parks 2015). • In this study we considered that greenways did not • We considered that greenways role in green have the amenities of parks but could be protected gentrification may be strongly dependent on city spaces (i.e. when a greenspace was both a greenway characteristics (residents’ commuting, city and a park was considered a park; but when it was structure). both a greenway and a nature preserve it was considered a nature preserve). M Triguero-Mas et al Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al specifically, we developed a model per city and gentri- in terms of the maximum values (pre ∼ 2000s max- fication time period so we had a maximum of three imum is 6042.86, while pre ∼ 2010s one is 1424.76). models per city: (a) one model with gentrification More specifically, when exploring the value of the in the decade of the 2000s as the outcome variable 95th percentiles, Lyon, Amsterdam and Valencia had and Parks pre ∼ 2000s, Gardens pre ∼ 2000s, Pre- the highest values of park area built pre ∼ 2000s, while serves pre ∼ 2000s, Recreational pre ∼ 2000s, and Nantes, Copenhagen and Amsterdam had them for Greenways pre ∼ 2000s as the exposure variables of the values pre ∼ 2010s. The rest of greenspace types interest; (b) one model with gentrification in the (gardens, nature preserves, recreational and green- decade of the 2010s as the outcome variable and ways) had 95% of their values of areas built below five Parks pre ∼ 2010s, Gardens pre ∼ 2010s, Preserves and maximum values below 2000. pre ∼ 2010s, Recreational pre ∼ 2010s, and Green- SGU average gentrification scores were generally ways pre ∼ 2010s as the exposure variables of interest; higher in the US, particularly in Baltimore, Detroit and (c) one model with gentrification considering the and Philadelphia in the 2010s. In contrast, they were decades of the 2000s and 2010s as the outcome vari- generally lower in Europe: Edinburgh in the 2000s, able and Parks pre ∼ 2000s, Gardens pre ∼ 2000s, Pre- Valencia in the 2000s and Vienna throughout the serves pre ∼ 2000s, Recreational pre ∼ 2000s, Green- 2000s and the 2010s (figure 3). ways pre ∼ 2000s as the exposure variable of interest). We reported the estimates of the associations 3.2. Analyses on gentrification and different types between gentrification and each independent vari- of new greenspaces altogether able using posterior probability values (postprob- Overall, when considering together mid-term gentri- value). These postprob-values indicate the probabil- fication (in the decade of the 2000s, in the decade of ity that each estimate that represents the effect of each the 2010s) and long-term gentrification (through the independent variable on gentrification is greater than decade of the 2000s and 2010s), we found associations zero. Note that this value is a measure of the relevance with several factors, with distance to the city center for each independent variable, so are not to be inter- and new parks being the most relevant ones (figure 4, preted in the same way as traditional p-values (sup- tables S5 and S6). More specifically, we found that— plemental material, page 23). in the presence of other covariates—new parks were positively associated with gentrification processes in more than half of our models. We also found sug- 3. Results gestive evidence that newly designated nature pre- serves were negatively associated with gentrification 3.1. SGU characteristics processes (i.e. negative links in half of our models). From the 28 cities included in our studies, those with Meanwhile, for new gardens, recreational spaces and SGUs with higher population density were those in greenways we did not find consistent results across the United Kingdom, Ireland and Nantes, while those cities. that extended more beyond the city center were all US cities (Austin, Detroit and Seattle) (table 2). In all our 3.3. Analyses on gentrification and different types cities, 50% of the SGU had not had any new public of new greenspaces separately by mid-term and transport stops built before the 2000s or before the long-term gentrification 2010s. The cities where a higher number of public Our findings of new parks being positively associ- transport stops had been built in a SGU before the ated with gentrification processes in more than half 2000s were all in Europe (Barcelona, Dublin, Lyon, of our models remained when considering both mid- Nantes and Sheffield) while before the 2010s the cities (in the decade of the 2000s and particularly in the with a higher number of public transport stops built decade of the 2010s) and long-term gentrification in a SGU were in Barcelona, Portland, San Francisco processes (figure 4, tables S5 and S6). For example, and Nantes. The cities where a high number of new in the cities of Atlanta, Boston, Denver, Louisville, residential buildings per SGU was sustained over time Milwaukee, Portland, San Francisco, Seattle, Van- were in North America (i.e. Portland and Vancouver). couver, Barcelona and Dublin our data showed links In contrast, the cities with an average higher SGU between new parks and both mid-term gentrification green coverage before the start of our study period and long-term gentrification processes. Moreover, in were in Europe (Barcelona, Copenhagen, Edinburgh, Austin, Cleveland, Washington D.C., Bristol, Copen- Valencia and Vienna). hagen, Nantes, and Vienna new parks were associated The average newly inaugurated and acquired with mid-term gentrification processes. greenspaces at the SGU level differed widely among However, the negative association between that greenspace types (figure 2). Parks were the spaces with newly designed nature preserves and gentrification the highest area built per SGU from all greenspace processes were more consistent for long-term gentri- types in terms of the 95th percentile (pre ∼ 2000s val- fication processes (figure 4, table S6) than for mid- ues being 95th percentile = 230.93 and pre ∼ 2010s term gentrification (table S5). Indeed, we found 95th percentile = 25.19, data not shown) and also that more area of newly designated nature preserves 7 Table 2. Descriptive statistics of census tract (SGU) characteristics, by city. Number of public transit stops added… New residential building built in the tract … Inhabitants per pre2000s pre2010s pre2000s pre2010s Green coverage square kilometer in Centre [metres: (stops/year: median (stops/year: median (buildings/year: (buildings/year: prior to 1990 [ha: 2010 [Inhab./ km2: mean (sd)] (95th pct)) (95th pct)) median (IQR)) median (IQR)) mean (sd)] mean (sd)] Atlanta 14 940 (9902) 0 (0.06) 0 (0) 0.02 (0.06) 0 (0.01) 0.04 (0.08) 1451 (1057) Austin 20 019 (14 261) 0 (0) 0 (0.13) 0.50 (5.70) 0.20 (4.60) 0.05 (0.09) 1258 (1285) Baltimore 4986 (2592) 0 (0.17) 0 (0) 0 (0) 0 (0) 0.25 (0.29) 4686 (3396) Environ. Res. Lett. 17 (2022) 104035 Boston 5615 (3382) 0 (0) 0 (0) 0 (0.01) 0 (0) 0.36 (0.47) 9449 (6744) Cleveland 12 359 (5858) 0 (0) 0 (0) 0.50 (1.90) 0 (0.20) 0.08 (0.17) 1978 (1193) Denver 7391 (4651) 0 (0.17) 0 (0.13) 1.40 (5.95) 0.60 (3.20) 0.13 (0.17) 2808 (1,534) Detroit 16 914 (8953) 0 (0) 0 (0) 0 (0.60) 0 (0) 0.05 (0.11) 1937 (1020) Louisville 11 763 (6450) — — 0 (0.10) 0.20 (0.75) 0.06 (0.12) 1448 (858) Milwaukee 8094 (4468) — — 0 (0.20) 0 (0) 0.01 (0.03) 3081 (2186) 8 Philadelphia 8236 (5316) 0 (0) 0 (0) 0.30 (2.30) 0.20 (1.00) 0.21 (0.28) 7430 (4188) Portland 9048 (6448) 0 (0.17) 0 (0.32) 4.25 (7.55) 4.00 (7.55) 0.07 (0.10) 2513 (1479) San Francisco 4875 (2815) 0 (0.17) 0 (0.38) 1.65 (2.95) 0.20 (0.80) 0.39 (0.47) 11 563 (8915) Seattle 17 582 (11 196) 0 (0) 0 (0.13) 0 (1.80) 0 (2.70) 0.06 (0.12) 2115 (2177) Washington DC 5250 (2419) 0 (0.08) 0 (0) 0 (0.01) 0 (0) 0.37 (0.34) 6195 (4398) Calgary 8553 (3817) 0 (0) 0 (0.14) 0 (1.10) 0.20 (7.05) 0.13 (0.23) 2708 (1579) Montreal 8710 (6001) 0 (0) 0 (0) 0.10 (0.30) 0 (0) 0.29 (0.28) 7833 (5472) Vancouver 4958 (2436) 0 (0.08) 0 (0.14) 5.65 (10.98) 5.60 (10.75) 0.18 (0.21) 7103 (5522) Amsterdam 3950 (2365) 0 (0) 0 (0) 0.33 (1.92) 0.13 (0.69) 0.38 (0.32) 10 468 (7627) Barcelona 3418 (1614) 0 (1.00) 0 (1.00) 0.10 (0.30) 0 (0) 0.98 (1.16) 43 103 (24 066) Bristol 3828 (2155) — — 2.22 (4.44) — 0.60 (0.39) 6078 (3496) Copenhagen 3301 (1652) — 0 (0) — 0 (0.17) 0.86 (1.22) 15 353 (11 298) Dublin 4085 (1885) 0 (0.43) 0 (0) — — 0.36 (0.26) 6512 (4273) Edinburgh 4387 (2861) 0 (0) — 0.33 (4.17) — 1.15 (0.99) 7627 (6278) Lyon 2493 (1249) 0 (0.33) 0 (0.14) — 2.64 (9.75) 0.58 (0.39) 15 756 (9990) Nantes 2712 (1331) 0 (0.44) 0 (0.22) — 11.57 (14.36) 0.16 (0.17) 6867 (3615) Sheffield 5025 (2817) 0 (0.33) — 1.11 (4.33) — 0.51 (0.40) 4833 (510) Valencia 2387 (1630) 0 (0.25) — 0.10 (0.30) — 0.94 (1.01) 31 189 (19 437) Vienna (central) 4543 (2673) — — — — 0.83 (0.73) 14 644 (11 051) Notes: Indicates that no data was available. Mean (standard deviation (sd)) is reported for continuous parametric variables. Median (interquartile range (IQR) is shown for continuous non-parametric variables, with the exception of those with a high percentage of zeros, for which median (95th percentile (95th pct)) is reported. M Triguero-Mas et al Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Figure 2. Descriptive statistics (median, 95th percentile) of new greenspace of different typologies before 2000s and also before 2010s, by city. White cells indicate data not available. For Vienna, nature preserves pre ∼ 2010s, recreational spaces pre ∼ 2010s and greenways pre ∼ 2010s only had one value different than zero each, and for that reason they were excluded from analyses, so their descriptives are not included. was associated with smaller long-term gentrification 3.4. Analyses on gentrification and different types indexes for all cities for which we had data except of new greenspaces over time periods for three: Boston (postprob-value = 0.70), Cal- When exploring if associations between new green- gary (postprob-value = 0.65) and Nantes (postprob- spaces and gentrification were sustained over differ- value = 0.79). ent time periods we did not find any clear patterns We also found that long-term gentrification was (tables S5 and S6). Only for parks, gardens and gre- positively associated with new gardens and greenways enways we did find sustained associations over time inaugurated between the start of our study period and in some cities. We found that parks pre ∼ 2000s were the 2000s in more than half of our models. Last, our positively linked with gentrification in the 2000s while results indicated some positive relationships between parks pre ∼ 2010s were associated with increased long-term gentrification and new recreational spaces gentrification in the SGU in the 2010s in Atlanta (in Atlanta, Baltimore, Denver, Washington D.C: and (postprob-value = 0.96 for the association with parks Barcelona) inaugurated between the start of our study pre ∼ 2000s and postprob-value = 0.99 for the associ- period and the 2000s (i.e. positive associations in ation with parks pre ∼ 2010s), Cleveland (postprob- half of the relationships we explored). In contrast, we value = 0.68 and postprob-value = 0.87), Denver found inconsistent results between new greenways, (postprob-value >0.99 and postprob-value = 0.66), recreational spaces and gardens and mid-term gentri- Louisville (postprob-value = 0.89 and postprob- fication indexes. value = 0.77), Seattle (postprob-value = 1.00 and 9 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Figure 3. Descriptive statistics (mean and standard deviation (sd)) of gentrification indicators, by city. White cells indicate data not available. postprob-value = 0.75) and Nantes (postprob- new parks and gentrification increasing over time value = 0.99 and postprob-value = 0.73). The same as—from the cities we included in the study— trend was found in Denver for greenways and in Mil- we found a higher proportion of the associations waukee, Seattle and Nantes for gardens. Moreover, we between parks and gentrification for gentrification in found some indications of the relationship between the 2010s than in the decade of the 2000s. 10 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Figure 4. Summary of the associations between different types of greenspaces (parks, gardens, nature preserves, recreational spaces, and greenways) and gentrification indexes (mid-term, long-term and considering mid and long-term gentrification altogether). The summary is presented for all the cities together and also for each geographical context (European, US, Canada). 3.5. Analyses on gentrification and different types included in our study, only in Baltimore, Detroit of new greenspaces by geographical area and Philadelphia our data did not reveal a positive Comparisons of our results by geographical area link between parks and gentrification. In contrast, we revealed that positive associations between new parks did not find any clear patterns for parks in Europe, and gentrification were predominantly found in the while in Canada we mostly found negative associ- US, both overall and when looking only at mid- or ations between parks and gentrification in most cities. long-term gentrification processes (figure 4, tables Similarly, our findings on newly designated S5 and S6). For example, from all the US cities we nature preserves being negatively linked with 11 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al gentrification, particularly when considering long- the associations and also due to our study avoiding term gentrification, were also mostly found in the reverse causation (i.e. our investigation estimating US, where from all the US cities for which we had the association between greenspace inaugurated or data, only in Boston we found links between newly acquired at a time period and gentrification in a designated nature preserves before the 2000s and later period, while the study by Rigolon & Németh gentrification processes in the decades 2000s–2010s. estimated the relationships between greenspace and Meanwhile, in Europe and Canada no consistent pat- simultaneously-occurring gentrification). terns were found for newly designated nature pre- Our results may indicate that the gentrification serves and gentrification processes. Last, the positive effects related to urban greening agendas and asso- relationships we found between new gardens and ciated green branding (Garcia-Lamarca et al 2019) gentrification were predominantly found in Canada may be strongly centered on parks. ‘Normative’ aes- and those for greenways were mostly found in Europe thetics and recreational possibilities may make parks and the US. the type of greenspaces that best encompass famili- arity, favorability and uniqueness and provide more 4. Discussion utility to neighborhood residents. All these factors are linked with green branding and gentrification Overall, this unique multi-city study on the rela- and may increase willingness to pay for these neigh- tionship between different types of greenspaces and borhoods (Chan 2015, Amorim Maia et al 2020). gentrification clearly demonstrates that—apart from For example, aspects such as ornate landscapes, with other commonly recognized factors such as distance manicured garden designs and fences (which might to the city center—different types of newly inaugur- give high feelings of security) could make parks ated or acquired greenspaces have a different role in more aestheticized and attractive to gentrifiers and the gentrification process. Through an international real estate developers (Chen et al 2021). They are analysis that reveals trends at the city-level based on also those spaces where racialized minorities might SGU data, we find that parks are the type of green- find themselves more controlled by police, surveilled, spaces more consistently positively associated with and criminalized (Anguelovski and Connolly 2021, gentrification processes. We also found some indic- Hoover and Lim 2021). ations of newly designated nature preserves being Moreover, the higher prevalence we found of pos- negatively associated with gentrification processes, itive relationships between new parks and gentrific- particularly when considering long-term gentrifica- ation in more recent times (i.e. from the cities we tion processes (i.e. throughout the 2000s and the included in the study, we found a higher propor- 2010s) and in the US. Meanwhile, for new gardens, tion of associations between new parks and gentrific- recreational spaces and greenways our research shows ation for gentrification in the 2010s than in the dec- mixed results. ade of the 2000s) are in line with findings from recent Our research indicates that different types of studies identifying the 2010s as the decade in which greenspace can be associated with gentrification cities have deployed a more articulated and formal (solely or together with other factors such as distance green branding (Gulsrud et al 2013, Andersson 2016, to the city center). This finding is aligned with most of Garcia-Lamarca et al 2019). This comprehensive the existing evidence (Lutzenhiser and Netusil 2001, green branding may be the result of developers and Czembrowski and Kronenberg 2016, Kwon et al 2017, cities around the world aiming at reproducing the Maantay and Maroko 2018, Rigolon and Németh marketing and private profit-making ‘success’ of New 2020, Chen et al 2021, Kim and Wu 2022). York’s High Line and Atlanta’s BeltLine (Loughran Our findings about parks as the greenspace type 2014, Haase et al 2017, Weber et al 2017, Immer- more consistently positively associated with gentrific- gluck and Balan 2018, García-Lamarca et al 2022). ation processes are partly in line with previous sci- The opening of both green interventions was linked entific evidence. A study in Barcelona had already to an increase in property values (and green gentrific- found that new parks built in the old town or in ation), which cities and developers potentially impor- traditionally industrialized neighborhoods explained ted to other contexts (Haase et al 2017, Rigolon and gentrification processes (Anguelovski et al 2018). Németh 2020). However, Rigolon & Németh did not find a link We are aware of only three previous studies on between acreage of new parks and gentrification in the impact of other typologies of greenspaces (such as their pooled analyses of data from ten US large and nature preserves, gardens, recreational spaces or gre- medium-sized cities (Rigolon and Németh 2020). The enways) on gentrification. A previous study demon- differences in results with our study could be due to strated that greenways created between 2008 and 2016 our research design including a more complete and increased the odds of census tract gentrification, but refined set of indicators to calculate gentrification those created between 2000 and 2008 were not linked scores, controlling of spatial autocorrelation rather to gentrification (Rigolon and Németh 2020). Other than using multilevel logistic regressions to estimate research found that the availability of community 12 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al gardens was modestly associated with gentrification contexts and time periods while considering the het- (Braswell 2018, Maantay and Maroko 2018). Our erogeneity of these different cities are strong elements findings add to the existing, yet still scarce, evid- of our study. Our study design also prevents reverse ence. The heterogeneity of associations we found may causation, as neighborhood gentrification cannot indicate that gardens, recreational spaces and green- casually influence new greenspace creation, inaugur- ways are types of greenspaces with very diverse access- ation, or improvement in previous years. ibility, uses, aesthetic, and functional characteristics Our study, however, also faces some limitations. and roles in gentrification processes. It may also be Our aim to compare a broad range of cities in dif- that other factors from which we did not have data ferent regions prevented us from using a completely (such as presence of vacant land) may influence the consistent data protocol, so we had to find comprom- role of community gardens, recreational spaces or ises in the use of time periods and variables across greenways in gentrification processes. the different cities. We use SGU as the unit of ana- Our findings that nature preserves are negatively lysis, but SGU limits do not determine gentrification associated with gentrification processes, particularly processes and we had low variability in our green- when considering long-time gentrification processes space variables (i.e. a lot of our census tracts had no or in US cities, are novel and have not been explored new greenspaces). Green gentrification may be bet- before. We hypothesize that most of the nature pre- ter assessed through analyses using greenspaces as the serves included in our analyses and categorized as unit of analysis, despite the high level of data pro- new greenspaces are in fact greenspaces that were cess that developing these measures can need may already present before our period of study (either make them inefficient at large scale. Finally, our study before the 1990s or before the 2000s), but which includes a very detailed dataset of types of new green- became enclosed, formalized as protected areas or spaces, but factors such as neighborhood historic sig- whose management changed from private to pub- nificance and characteristics, greenspace size or pres- lic during our study period (1990–2016). These legal ence of private greenspaces that may be correlated or administrative changes potentially may not trans- with both gentrification and public greenspaces, and late into new value creation. An alternative explan- that previous studies have explored as explanatory ation could be that preserves may not be perceived variables (Anguelovski et al 2018, Chen et al 2021, nor allowed (like La Petite Amazonie in Nantes) as Donovan et al 2021, Kim and Wu 2022) were not formal accessible green spaces for daily use or may included. even be associated with feelings of fear or insecurity for some people due to their wildness (Lyytimäki and Sipilä 2009, Rupprecht and Byrne 2014). 5. Conclusions The differences we find between the US and Canada and Europe in the occurrence of green gentri- In summary, our study reveals that parks—together fication may also be due to distinct policy and plan- with other factors such as proximity to the city ning contexts, with European cities generally bene- center—may contribute to gentrification processes fiting from the greater presence of anti-displacement (i.e. positive associations in more than half of our tools and equitable greening practices (Oscilowicz models), particularly in the US context, except in his- et al 2022) in comparison with US cities. Last, the torically black disinvested postindustrial cities with importance of spatial effects found in our analyses an abundance of vacant land. In contrast, nature pre- indicates that new greenspaces located near gentrify- serves, greenways, gardens and recreational spaces are ing neighborhoods are more susceptible to gentrifica- types of greenspaces with diverse roles in gentrific- tion than those located far away, which is in line with ation processes (some positive, some negative and previous evidence (Pearsall and Eller 2020). This may some no effect associations). We conclude that the suggest that greenspaces serve as an anchor to support phenomenon of green gentrification in North Amer- gentrification processes already existing in the vicin- ica in particular, and in Europe to some extent may ity, but do not have an impact alone. be reinforced more by parks than other greenspaces Our original and methodologically robust study types. explores the associations between a broad range of Both gentrification and an increase in re-naturing types of greenspaces and gentrification. To our know- projects are well-described in cities around the world ledge, it is the first to investigate these associations (Lees 2012, Haase et al 2017, Navarrete Escobedo in a broad range of cities in Europe, Canada and 2020). Moreover, green gentrification is growingly the US, across different time periods using rigorous reported in different locations (Quinton et al 2022). novel methods, namely a Bayesian spatial analysis Accordingly, although our findings refer to Europe approach, to do so. More specifically, our control of and North America, they should be considered in spatial autocorrelation and a broad range of potential other contexts and future studies should explore if our confounders and our gentrification assessment with a results are replicated in other geographical locations score developed to be valid in different geographical such as South America or Asia. 13 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Future studies should further explore the poten- of Science, Innovation and Universities for PID2020- tial impact of different characteristics and other 115882RB-I00 grant. attributes of greenspace on gentrification, build- ing on the attributes already explored to produce ORCID iDs comparable evidence. When doing so, it will be important to investigate a combination of types and Margarita Triguero-Mas  https://orcid.org/0000- biophysical characteristics (e.g. parks of different 0002-1580-2693 size, parks with different attributes), including socio- Nick Martin  https://orcid.org/0000-0001-9023- cultural traits related to greenspaces attractiveness 9696 and perception (van Vliet et al 2020, Kim and Wu Francesc Baró  https://orcid.org/0000-0002-0145- 2022) and to use greenspace as the unit of analysis 6320 (as previously done by (Anguelovski et al 2018). The main objective of our study was to investigate whether References different types of greenspaces impact gentrification. However, as previous studies have suggested, future Amorim Maia A T, Calcagni F, Connolly J J T, Anguelovski I and lines of research should also explore if the causal Langemeyer J 2020 Hidden drivers of social injustice: uncovering unequal cultural ecosystem services behind link is inversed (i.e. if gentrification processes explain green gentrification Environ. Sci. Policy 112 254–63 the creation and designation of new greenspaces) Andersson I 2016 Green cities’ going greener? Local (Shokry et al 2020, Reibel et al 2021). environmental policy-making and place branding in the Considering the environmental and health bene- ‘Greenest City in Europe Eur. Plan. Stud. 24 1197–215 Anguelovski I 2016 From toxic sites to parks as (Green) LULUs? fits of urban re-naturing projects, we are calling for New challenges of inequity, privilege, gentrification, and cities to keep investing in improving park access and exclusion for urban environmental justice J. Plan. Lit. availability in under-privileged neighborhoods while 31 23–36 simultaneously implementing anti-displacement and Anguelovski I et al 2022 Green gentrification in European and North American cities Nat. Commun. 13 3816 green inclusive policies (Oscilowicz et al 2022). With- Anguelovski I, Connolly J J T, Masip L and Pearsall H 2018 holding greenspace investment and associated bene- Assessing green gentrification in historically disenfranchised fits would indeed risk reinforcing the health inequities neighborhoods: a longitudinal and spatial analysis of and environmental injustices that socially underpriv- Barcelona Urban Geogr. 39 458–91 Anguelovski I and Connolly J J (eds) 2021 The Green City and ileged residents already face. Social Injustice: 21 Tales from North America and Europe (New York: Routledge) (available at: www.routledge.com/ The-Green-City-and-Social-Injustice-21-Tales-from-North- Data availability statement America-and-Europe/Anguelovski-Connolly/p/book/ 9781032024110) The data that support the findings of this study are Braswell T H 2018 Fresh food, new faces: community gardening as available upon reasonable request from the authors. ecological gentrification in St. Louis, Missouri Agric. Hum. Values 35 809–22 Brown-Saracino J 2010 The Gentrification Debates. A Reader (London: Taylor & Francis) Acknowledgments Chan C S 2015 Public parks in city branding: perceptions of visitors Vis-à-vis residents in Hong Kong Urban For. Urban The research presented in this paper received funding Green. 14 1157–65 from the European Research Council (ERC) under Chen Y, Xu Z, Byrne J, Xu T, Wang S and Wu J 2021 Can smaller parks limit green gentrification? Insights from Hangzhou, the European Union’s Horizon 2020 research and China Urban For. Urban Green. 59 127009 innovation program [grant agreement No. 678034], Connolly J J T 2019 From Jacobs to the Just City: a foundation for the Spanish Ministry of Science and Innovation challenging the green planning orthodoxy Cities 91 64–70 [Maria de Maeztu, CEX2019-000940-M) and the Connolly J J T and Anguelovski I 2021 Three histories of greening and whiteness in American Cities Front. Ecol. Evol. 9 621783 Research Council VUB [SRP 16 Demographic chal- Conway D, Li C Q, Wolch J, Kahle C and Jerrett M 2010 A spatial lenges of the 21st century] but the sponsors had no autocorrelation approach for examining the effects of urban role in the design or analysis of this study. MGL greenspace on residential property values J. Real Estate and LA are funded by Juan de la Cierva fellowships Finance Econ. 41 150–69 Curran W and Hamilton T 2012 Just green enough: contesting [IJC2020-046064-I, and IJC2020-045101-I] awarded environmental gentrification in Greenpoint, Brooklyn Local by the Spanish Ministry of Economy and Competit- Environ. 17 1027–42 iveness. HVSC is supported by the Banco Santander- Cvejic R, Eler K, Pintar M, Zeleznikar S, Haase D, Kabisch N and UAB fellowship program. DC would like to thank Strohbach M 2015 A typology of urban green spaces, ecosystem services provisioning services and demands the Spanish Ministerio de Ciencia e Innovación— [Greensurge D3.1 report] Agencia Estatal de Investigación for grant PID2019- Czembrowski P and Kronenberg J 2016 Hedonic pricing and 106341GB-I00 (jointly financed by the European different urban green space types and sizes: insights into the Regional Development Fund, FEDER). BS was sup- discussion on valuing ecosystem services Landsc. Urban Plan. 146 11–19 ported by Margarita Salas fellowship from the Span- de Oliveira F L and Mell I (eds) 2019 Planning Cities with Nature: ish Ministry of Universities-University of Valencia Theories, Strategies and Methods (Cham: Springer) (https:// (MS21-013). JM-M would like to thank the Ministry doi.org/10.1007/978-3-030-01866-5) 14 Environ. Res. Lett. 17 (2022) 104035 M Triguero-Mas et al Ding L, Hwang J and Divringi E 2016 Gentrification and Oscilowicz E, Anguelovski I, Triguero-Mas M, García-Lamarca M, residential mobility in Philadelphia Reg. Sci. Urban Econ. Baró F and Cole H V S 2022 Green justice through policy 61 38–51 and practice: a call for further research into tools that foster Donovan G H, Prestemon J P, Butry D T, Kaminski A R and healthy green cities for all Cities Health (https://doi.org/ Monleon V J 2021 The politics of urban trees: tree planting 10.1080/23748834.2022.2072057) is associated with gentrification in Portland, Oregon For. Pearsall H and Eller J K 2020 Locating the green space paradox: a Policy Econ. 124 102387 study of gentrification and public green space accessibility in Dooling S 2009 Ecological gentrification: a research agenda Philadelphia, Pennsylvania Landsc. Urban Plan. 195 103708 exploring justice in the city Int. J. Urban Reg. Res. 33 621–39 Quinton J, Nesbitt L and Sax D 2022 How well do we know green Du M and Zhang X 2020 Urban greening: a new paradox of gentrification? A systematic review of the methods Prog. economic or social sustainability? Land Use Policy 92 104487 Hum. Geogr. 46 960–87 Freeman L 2005 Displacement or succession?: Residential mobility Reibel M, Rigolon A and Rocha A 2021 Follow the money: do in gentrifying neighborhoods Urban Aff. Rev. 40 463–91 gentrifying and at-risk neighborhoods attract more park García-Lamarca M, Anguelovski I, Cole H V S, Connolly J J T, spending? J. Urban Aff. (https://doi.org/10.1080/07352 Pérez-del-Pulgar C, Shokry G and Triguero-Mas M 2022 166.2021.1886857) Urban green grabbing: residential real estate developers Rigolon A and Németh J 2018 Privately owned parks in New discourse and practice in gentrifying Global North Urbanist communities: a study of environmental privilege, neighborhoods Geoforum 128 1–10 equity, and inclusion J. Urban Aff. 40 543–59 García-Lamarca M, Anguelovski I, Cole H, Connolly J J, Rigolon A and Németh J 2020 Green gentrification or ‘just green Argüelles L, Baró F, Loveless S, Del Pulgar Frowein C P and enough’: do park location, size and function affect whether a Shokry G 2019 Urban green boosterism and city place gentrifies or not? Urban Stud. 57 402–20 affordability: for whom is the ‘branded’ green city? Urban Rupprecht C D D and Byrne J A 2014 Informal urban greenspace: Stud. 58 004209801988533 a typology and trilingual systematic review of its role for Glass R 1964 London: Aspects of change Aspects of Change urban residents and trends in the literature Urban For. (MacGibbon & Kee) p 13 Urban Green. 13 597–611 Gould K A and Lewis T L 2017 Green Gentrification: Urban Safransky S 2014 Greening the urban frontier: race, property, and Sustainability and the Struggle for Environmental Justice resettlement in Detroit Geoforum 56 237–48 (London: Taylor & Francis) Sandberg L A 2014 Environmental gentrification in a Gulsrud N M, Gooding S and van den Bosch C C K 2013 Green post-industrial landscape: the case of the Limhamn quarry, space branding in Denmark in an era of neoliberal Malmö, Sweden Local Environ. 19 1068–85 governance Urban For. Urban Green. 12 330–7 Schinasi L H, Cole H V S, Hirsch J A, Hamra G B, Gullon P, Haase D et al 2017 Greening cities—to be socially inclusive? Bayer F, Melly S J, Neckerman K M, Clougherty J E and About the alleged paradox of society and ecology in cities Lovasi G S 2021 Associations between greenspace and Habitat. Int. 64 41–48 gentrification-related sociodemographic and housing cost Hoover F A and Lim T C 2021 Examining privilege and power in changes in major metropolitan areas across the United US urban parks and open space during the double crises of States Int. J. Environ. Res. Public Health 18 3315 antiblack racism and COVID-19 Soc. Ecol. Pract. Res. Shokry G, Connolly J J and Anguelovski I 2020 Understanding 3 55–70 climate gentrification and shifting landscapes of protection Immergluck D 2009 Large redevelopment initiatives, housing and vulnerability in green resilient Philadelphia Urban Clim. values and gentrification: the case of the Atlanta Beltline 31 100539 Urban Stud. 46 1723–45 Smith G, Cirach M, Swart W, Dėdelė A, Gidlow C, van Kempen E, Immergluck D and Balan T 2018 Sustainable for whom? Green Kruize H, Gražulevičienė R and Nieuwenhuijsen M J 2017 urban development, environmental gentrification, and the Characterisation of the natural environment: quantitative Atlanta Beltline Urban Geogr. 39 546–62 indicators across Europe Int. J. Health Geogr. 16 1 Irwin E G 2002 The effects of open space on residential property Stuhlmacher M, Kim Y and Kim J E 2022 The role of green space values Land Econ. 78 465–80 in Chicago’s gentrification Urban For. Urban Green. Jo Black K and Richards M 2020 Eco-gentrification and who 71 127569 benefits from urban green amenities: NYC’s high Line Taylor L and Hochuli D F 2017 Defining greenspace: multiple uses Landsc. Urban Plan. 204 103900 across multiple disciplines Landsc. Urban Plan. 158 25–38 Kabisch N, Korn H, Stadler J and Bonn A (eds) 2017 Nature-Based Triguero-Mas M, Anguelovski I, García-Lamarca M, Argüelles L, Solutions to Climate Change Adaptation in Urban Areas: Perez-del-pulgar C, Shokry G, Connolly J J T and Linkages between Science, Policy and Practice (Cham: Cole H V S 2021 Natural outdoor environments’ health Springer) (https://doi.org/10.1007/978-3-319-56091-5) effects in gentrifying neighborhoods: disruptive green Kim S K and Wu L 2022 Do the characteristics of new green space landscapes for underprivileged neighborhood residents Soc. contribute to gentrification? Urban Stud. 59 360–80 Sci. Med. 279 113964 Kwon Y, Joo S, Han S and Park C 2017 Mapping the distribution van Vliet E, Dane G, Weijs-Perrée M, van Leeuwen E, van pattern of gentrification near urban parks in the case of Dinter M, van den Berg P, Borgers A and Chamilothori K Gyeongui Line Forest Park, Seoul, Korea Sustainability 9 231 2020 The influence of urban park attributes on user Lees L 2012 The geography of gentrification: thinking through preferences: evaluation of virtual parks in an online comparative urbanism Prog. Hum. Geogr. 36 155–71 stated-choice experiment Int. J. Environ. Res. Public Health Loughran K 2014 Parks for profit: the high line, growth machines, 18 212 and the uneven development of urban public spaces: parks Veerkamp C J, Schipper A M, Hedlund K, Lazarova T, Nordin A for profit City Commun. 13 49–68 and Hanson H I 2021 A review of studies assessing Lutzenhiser M and Netusil N R 2001 The effect of open spaces on ecosystem services provided by urban green and blue a home’s sale price Contemp. Econ. Policy 19 291–8 infrastructure Ecosyst. Ser. 52 101367 Lyytimäki J and Sipilä M 2009 Hopping on one leg—the challenge Weber S, Boley B B, Palardy N and Gaither C J 2017 The impact of of ecosystem disservices for urban green management urban greenways on residential concerns: findings from the Urban For. Urban Green. 8 309–15 Atlanta BeltLine Trail Landsc. Urban Plan. 167 147–56 Maantay J and Maroko A 2018 Brownfields to greenfields: Wolch J R, Byrne J and Newell J P 2014 Urban green space, public environmental justice versus environmental gentrification health, and environmental justice: the challenge of making Int. J. Environ. Res. Public Health 15 2233 cities ‘just green enough’ Landsc. Urban Plan. 125 234–44 Navarrete Escobedo D 2020 Foreigners as gentrifiers and tourists Yang B Y et al 2021 Greenspace and human health: an umbrella in a Mexican historic district Urban Stud. 57 3151–68 review Innovation 2 100164 15