Urban Public Space

Every day society travels through different spaces that make up the city and others that serve for our transition: free transit sites, restricted access ones; public spaces, private ones; covered, semi covered, outdoor sites; walkways, vehicular roads, gardens and buildings.

Given this diversity of spaces and temporalities, public space is very difficult to quantify. Delimiting it geometrically in the urban plan is quite a methodological challenge (^{Ji & Ding, 2021}). As emphasized by ^{Barchetta and Chiodelli (2015)}, there is a multiplicity of institutional arrangements behind public spaces: combinations of property regimes, forms of management, maintenance, and behavioral norms that create de jure and de facto access restrictions (intentional or not).

The following is an example of a detailed work of delineation of the public space:

in 1736, Pope Clement XII commissioned Giambattista Nolli, an Italian architect and surveyor, to draw an accurate map of Rome. For public buildings related to religion, culture, and city administration, such as churches and theaters, the interior space was left blank, and the columns and walls were painted in detail. In addition to the public buildings, the map also illustrates the public parts of the private buildings, such as the atrium and the public corridors that connect the entrance to the inner courtyard. In other words, the white areas on the map represent the areas of the city that are actually accessible and reflect the areas of public life and resident activity. The private space represents places that people cannot enter and is filled in with dark gray shading (^{Ji & Ding, 2021, p. 542}).

Indicator 11.7.1 of the 2030 Agenda tries, with modern tools, to replicate what Giambattista Nolli did in detail in 1736 for the city of Rome. This indicator imposes significant challenges by requiring spatial data and technical capacities for its analysis. The indicator has two dimensions: a) the geophysical, which consists in obtaining the public space of the urban fabric; and b) the social, which determines the use and access of people by gender, age and disability. Regarding the difficulty in estimating the indicator, ^{Cramwinckel (2019)} argues that, even though it is conceptually clear and has a standard international methodology, data has not yet been collected.

To have an overview of what is happening in Mexico, UN-Habitat recommends that only 30-35% of the total OPS be for streets and sidewalks. In 2020, the UN-Habitat methodology was applied in 23 cities in Mexico, not including Hermosillo. The averages obtained were below the recommended ranges: streets and sidewalks 13.28%, OPS 2.51% (^{UN-Habitat, 2021}). By adopting indicator 11.7.1, urban parks would be included in the OPS percentage, which represents an extremely low value.

Urban Parks

There is a wide range of definitions of green spaces, both among academics and in legal texts and official documents (^{Canosa et al., 2003}; ^{Hartig et al., 2014:211}; ^{Barchetta & Chiodelli, 2015}; ^{Téllez Montes et al., 2020}; ^{NOM-001-SEDATU-2021 de 2022}). ^{Tellez-Montes et al. (2020)} point out inconsistencies in the determination and categorization of green spaces in Mexico City. The authors suggest that it is necessary to start from clear concepts in order to create and apply consistent, clear and functional public policies to protect and promote these spaces.

There is also much confusion between concepts such as ecosystem, vegetation cover on a city, neighborhood and block scale, tree cover or urban forest. This confusion makes the construction of indicators even more difficult (^{López González et al., 2021}).

In this work we use the term “urban park” to designate elements that are a subset of public space, with free access, connected (or overlapping) at a walkable distance from the urban sprawl, with different proportions of vegetation cover (tree, shrub or herbaceous), deliberately designed and built with equipment to carry out outdoor activities, leisure, recreation, rest, social interaction, enjoyment of environmental amenities, sports, among others.

Recent reforms to the Land Planning and Urban Development Law of the State of Sonora [Ley de Ordenamiento Territorial y Desarrollo Urbano del Estado de Sonora] (^{Decree 144 of 2020}) embody this idea, defining green public space as a plaza or green corridor with public access, also establishing that: “the ridges, sidewalks and roundabouts devoid of social infrastructure (benches, walkways, playsets) must be excluded” from this category, “as well as those areas not accessible to the entire population” (p. 5).

In this definition we place an anthropocentric and salutogenic emphasis, understanding the urban park as a promoter and an asset for the well-being and health of people. The WHO produced a report summarizing the benefits that green spaces offer to cities. In general, they improve mental health, reduce both morbidity and mortality from cardiovascular diseases, and contribute to reducing the rates of obesity and type 2 diabetes. They also help to have better pregnancies. All these benefits are mediated by psychological relaxation and stress relief. Citizens can increase their physical activity and be less exposed to air pollutants, noise and excess heat (^{WHO, 2016}).

Figure 1 shows how these three categories often create confusion: green spaces, urban parks and indicator 11.7.1 of the 2030 Agenda. First, as already mentioned, this indicator seeks to measure the entire public space. Second, in the urban sprawl there are areas of cultivated or spontaneous vegetation (undergrowth or remnants of native vegetation) that provide greenery, however, only part of this vegetation is found in public spaces. Additionally, in arid areas such as Hermosillo there are parks that have very little or no vegetation, projecting a brown or grayish view (^{Clemente Marroquín, 2007}; ^{Bernal Grijalva et al., 2019}). Third, there are also non-public parks, such as those located inside condominiums with exclusive access for their residents. Finally, in figure 1 the intersection of the three circles suggests that ideally an urban park should be public, have a legal use defined as a park and have adequate green cover.

Source: Own elaboration based on the reviewed literature (^{Barchetta & Chiodelli, 2015}; ^{López González et al., 2021}; ^{NOM-001-SEDATU-2021 of 2022}).

Figure 1. Venn diagram with the interaction of park, public space and green areaNote: NDVI: Normalized Difference Vegetation Index; SEDATU: Secretaría de Desarrollo Agrario, Territorial y Urbano, México (The Mexican Secretariat for Agrarian, Land and Urban Development).  

Integral System of Urban Parks

On an urban scale, parks should not be managed in isolation. It is suggested that a comprehensive system of urban parks should be built (^{Harnik, 2006}; ^{Peña Salmón, 2011}; ^{Ellis & Schwartz, 2016}; ^{NOM-001-SEDATU-2021 de 2022}). The set of spaces considered as parks within an urban municipality is defined as a system of urban parks; basically, this system ensures the fair distribution of the recreational, cultural, and environmental amenities provided by these spaces (^{Ellis & Schwartz, 2016}).

^{Corbett and Mellouli (2017)} studied the importance of information systems to support the monitoring and achievement of the goals and targets of the 2030 Agenda. They highlighted that governments generate information systems oriented to daily administrative functions. However, they suggest that a good information system should be useful in three spheres: a) the political, related to the discourse and public agenda of elected politicians; b) the administrative, already mentioned; and c) that of sustainability, where the information is used to estimate the indicators of the 2030 Agenda with technical, academic and scientific support.

^{Harnik (2006)} highlights that, despite the importance of dimensioning and comprehensively knowing the parks located in a city, few cities have carried out this task. The author recommends seeking seven aspects to have a park system of excellence: a) establish and follow a clear mission and vision since the system does not emerge from nothing; b) prepare a master plan with citizen participation and keep it updated; c) allocate sufficient budget; d) distribute the parks homogeneously in the urban sprawl; e) monitor the satisfaction of park users; f) that they are safe spaces, free of crime and violence; and g) that they generate environmental, economic and social benefits.

^{Peña Salmón (2011)} developed a methodological proposal for the planning of urban green spaces for the city of Mexicali, Baja California. Peña subdivided the process into a) organization: definition of objectives and integration of the team; b) inventory: not only of parks, but of resources available to work with the methodology; c) integration of a diagnosis; d) strategy: programs and goals; and e) management: annual action plans. He also recommends and proposes creating an inventory of existing and potential green spaces based on typologies and preparing a fact sheet for each park.

A comprehensive system of urban parks plays a fundamental role from the perspective of environmental equity. In Mexico, the Political Constitution establishes that “everyone has the right to an adequate environment for their development and well-being” (^{Decree of 1999, Art. 4 o, para. 3}). The socio-spatial inequalities associated with the distribution of urban parks have been analyzed in several cities around the world, for example: Hermosillo (^{Lara-Valencia & García-Pérez, 2013}; ^{García Pérez & Lara-Valencia, 2016}), Mexico City (^{Fernández-Álvarez, 2017}), various cities in the United States (^{Boone et al., 2019}; ^{Nesbitt et al., 2019}; ^{Williams et al., 2020}), and Quito (^{Cuvi & Gómez-Vélez, 2021}). Said studies consistently find an unfair distribution of environmental amenities.

Finally, the recently published ^{NOM-001-SEDATU-2021 of 2022} advises that public spaces must be seen as an interconnected system or set and not be destinations in themselves.

The DPJ List

The DPJ did not have a map to locate and measure the areas under its responsibility (research question 1). Instead, they shared a list that had 1 020 rows, corresponding to 146 boulevards, 75 in rural areas and 799 squares. Rural areas and green spaces on boulevards were not included in this study, so the universe to be located and photo-interpreted was left at 799.

The DPJ does not publish or maintain a chronologically ordered registry or database that allows to know the evolution of the parks, boulevards and other areas under its responsibility. However, it was possible to have access to their 2016 and 2018 databases, which allowed a line-by-line comparison for the squares to be made and to know their changes in the last five years (see Table 1).

In an interview, a DPJ official pointed out that the list is dynamic and is constantly being updated. This list is divided into sectors and for each one there is a manager who mobilizes the maintenance crews. In addition, they are the ones who contribute to feeding the list based on their field work, determine the location and make the measurements of the place. The interviewed official also specified that by February 2022, 75 rural areas (391 264 m²), 152 boulevards (1 289 021 m²) and 814 parks (3 807 812 m²) were tended to. With this information, it can be deduced that from January 2016 to February 2022, the DPJ oversaw 145 more urban parks.

For the ^{DPJ list (2020)}, a digital survey was carried out using the photo-interpretation of ^{Google Earth images (2022)}. The resolution of the available images made it possible to distinguish various elements of the typical equipment of an urban park, such as shade structures, kiosks, sheds, roofs, basketball courts, walkways, sidewalks, children’s playsets and sports equipment, tree canopy, shrubs, synthetic grass, bare ground etc. This technique was very precise. To verify this, a random sample of one hundred spaces was taken for either a field visit or viewed on “street view”, and there was a 100% coincidence between the photo-interpretation and the elements seen in reality.

Out of 799 parks (2 975 638 m²), it was possible to locate and draw individual polygons for 700 sites totaling an area of 1 724 656 m² (see map 2), that is, in round numbers, only 58% of the inventoried area for the concept “square” or urban park. Table 2 details the findings of the photo-interpretation process of the ^{DPJ 2020} list.

Source: Own elaboration based on the ^{DPJ (2020)} list, ^{Google Earth (2022)} and ^{PlanetScope (2022)}.

Map 2. Centroids of the 700 polygons located (photo-interpreted) from the ^{DPJ list (2020)} 

The photo-interpretation presents some difficulties: 1) it’s a very time-consuming technique; 2) it’s manual and although it produces accurate results, it depends on the skills of the photo-interpreter; 3) regarding the ^{Google Earth platform (2022)}, it made up a mosaic of images from different dates, from November 2015 to June 2019; and 4) skills and abilities are required for the use of GIS software.

The diversity of surfaces that theoretically could be used in the numerator of the WHO m2/inhabitant indicator or, where appropriate, to analyze the spatial distribution of parks stands out. ^{Saldaña Vázquez and Ramírez Miranda (2021)}, referring to ^{Sánchez Juárez (2021)}, observed that he had included in his calculation spaces that are not easily accessible and not used for recreation, such as ridges, cemeteries or private green spaces. However, other authors included ridges and it would seem that they only include surface areas (of parks) with vegetation cover (^{Pérez-Medina & López-Falfán, 2015}).

^{García Pérez and Lara Valencia (2016, p. 35)}, interpreting state law⁶ as an “urbanized area”, estimated that the city of Hermosillo should have at least 6.9 m² of green space per inhabitant. However, the interpretation is wrong, the law uses the salable area as a basis for the creation of green spaces. Recalculating would yield an area of less than 6.9 m². This demonstrates the variety of forms of interpretation and operationalization of green spaces.

In the ^{DPJ list (2020)} there are many parks without a location and cemeteries considered as parks. ^{Quinton and Duinker (2019)} state that cemeteries are generally omitted as green spaces, even though they could have as much vegetation as a park and therefore generate environmental services. It is assumed that cemeteries must have considerable vegetation, which is not always the case, especially in Hermosillo.⁷ In addition, ^{Quinton and Duinker (2019)} allude that due to the influence of the media and popular culture, these spaces are considered gloomy and grim. In the social imagination they are associated with supernatural phenomena, horror stories and movies.

Far from the conceptual debate of whether or not to include cemeteries as “green spaces”, the reason for their inclusion is more a reason of administrative inertia or tradition. In the interview with a former director of the DPJ, he pointed out that “the cemeteries of Hermosillo have traditionally depended on the DPJ for their maintenance, especially, prior to the commemorative days they are cleaned, since these places do not have personnel to perform these tasks” (official 4, personal communication, January 28, 2022).

Regarding the unlocated spaces that totaled 402 305 m² (item B, table 2), a DPJ official noted that:

The Los Arroyos sports complex (30 hectares) began to be tended to, there was a sports league there, trees were planted, it was cleaned from time to time, but the league then changed locations, the land was private property (...) they were added [the places and area] so that it would reflect the work carried out by the DPJ (official 3, personal communication, February 9, 2022).

He was also asked about the “Akiwiki #2” case with 6 000 m², noting that “3 000 m² were removed that turned out to be private property.” The interviewee mentioned that “other places did not exist or were duplicated” (official 3, personal communication, February 9, 2022).

Another official explained the existence of private property within the DPJ inventory:

there are people who have land in the urban sprawl that is used as sports areas, “I have a large piece of land, I lend it to the municipality to use it as a sports area, so I have two benefits, that they don’t invade it, why? How are they going to invade a soccer field? and I pay half of the property tax”. But an agreement has to be made through the Municipal Sports Institute and the Treasury (Instituto del Deporte Municipal y la Tesorería) (official 5, personal communication, November 5, 2021).

Derived from the interviews given by three officials, the list of areas served by the DPJ increased: 1) due to political commitments, for example, the city council is asked for help to clean a park or sports area; 2) by administrative tradition, that is, if the municipality does not do it, who does?; 3) due to the need to report work based on maintenance and cleaning in square meters; and 4) by the delivery-reception from the developer to CIDUE, and the latter to the DPJ.

Likewise, areas are removed from the list for various reasons: 1) as cleanup and to detect registry errors; 2) green and sports areas that cease to function as such are taken into account; 3) recognition of areas that happen to be private property or intended for another use; and 4) the political commitment or agreement/request for help ends, generally made by the users of the park/sports area to the mayor in office.

Available Databases

This section answers the second research question and refers to the characteristics of the databases available to integrate a GIS of urban parks.

The first difficulty an academic faces when generating a complete and exhaustive inventory of urban parks is the lack of information. For this research it was not possible to obtain the raw data related to this topic in an editable format. It was found that between groups of experts, interested parties and/or consultants, data is shared in an editable format, although many times without “metadata” detailing its quality, update date, definition of attributes, etc. These support networks believe that this information should be officially available.

Table 3 summarizes the databases and digital cartographies consulted. None of the databases could be considered a complete and comprehensive inventory of green spaces or urban parks in Hermosillo (research question 1). It can also be observed that there is a temporal discontinuity between the sources of information, that is, there is no way to have serial observations that are like a “snapshot” taken at regular intervals that allow knowing the evolution of the variables over time.

An interviewed former Implan official stated: “we have perfectly geolocated all the green spaces of the city” (official 6, personal communication, October 21, 2021). A copy of these georeferences in digital format was requested from the Implan reception window for review, which was denied.

For the general public there is the 2014 version (^{Implan Hermosillo, 2014}) of the E6 plan, in its printed edition or in PDF. The public information was georeferenced and vectorized, however, the adjustment had discrepancies and the available format presented analytical limitations. A more current version of the E6 plan is the SIGEM interactive map, but it can only be consulted in real time through the internet (^{H. Ayuntamiento de Hermosillo, 2022}).

A former Implan official, as well as an active member of its advisory council, announced that the new Urban Development Plan of the Population Center of Hermosillo 2020-2050 (Plan de Desarrollo Urbano del Centro de Población de Hermosillo 2020-2050), will have a more detailed description of urban green spaces, which they call “system of urban green spaces”. However, this document is not yet available to the public (official 6, personal communication, October 21, 2021).

Geospatial Analysis

To find out if the 700 photo-interpreted parks on the ^{DPJ list (2020)} could be considered a complete and exhaustive inventory, various geospatial analyses were carried out to overlay them with other geographic databases.

1) It was identified that 185 parks had an area of less than 500 m². The PDUCPH Modification 2014 (^{Implan Hermosillo, 2014}) stipulated that green spaces should not have a surface area of less than 500 m²; nor can ridges, ravines, streams, hills and decorative surfaces of residential communities be added. Only 35 parks had a surface area greater than one hectare (range 1.5-9.1 hectares). The vast majority were narrow polygons; 232 with an average width of less than 20 meters (see graph 2).

Source: Own elaboration based on the list of the ^{DPJ (2020)} and ^{Google Earth (2022)}.

Graph 2. Histograms with the average width and the area of the photo-interpreted polygons 

2) Six hundred three parks with street views were located. The fact that not all the parks were located with Google Maps is not just a technical issue of the company; if the vehicle did not take a street view, most of the time it is due to some access restriction, as in the case of gated communities. Thirty of these sites were randomly inspected and in all cases the presence of security booths, fences and/or boom barriers for access control was verified. Strictly speaking, parks located within gated communities are not accessible to the public. In the opinion of a former official of the defunct Maintenance and Forestry Department of Hermosillo (Dirección de Mantenimiento y Forestación de Hermosillo), “there are gated communities of a certain [economic] level where they have to do it [maintenance] with their own resources” (official 4, personal communication, January 28, 2022); but as it was detected in this study, some of the parks are donations, owned by the municipality, which remain within gated communities.⁸

3) The geographic service coverage of the DPJ can be corroborated by superimposing the centroids of the photo-interpreted parks on the polygons of the Hermosillo neighborhoods as of 2019. The result was that 474 neighborhoods (9 900 hectares), out of a total of 776, did not have a park tended by the DPJ.⁹

4) Comparing the photo-interpreted parks with the Implan cartography (Plan E6), only 502 parks had “green and/or sports area” (AVD) as land use. Plan E6 contained a total of 2 285 polygons (886.345 hectares). Around 466.427 hectares were located on subdivided rural land, still without construction at the date of this investigation. Also noteworthy are the Los Lagos golf club (71.760 hectares), closed sports areas (166.401 hectares), abandoned parks such as La Sauceda (40 hectares) and some vacant land. Also, 1 047 polygons (20.023 hectares) have an area of less than 500 m², where flowerbeds, triangles, roundabouts, etc. can be seen, probably ornamental surfaces of residential communities. This analysis questions the precision and exhaustiveness of the E6 plan of the Implan, as a possible inventory of urban parks.

Finally, the geographic overlap of the parks with digital cadastre cartography for the year 2019 was analyzed. There was an overlap of 654 parks from different owners; 544 belong to the municipality, 31 to the state entity, four are federal and the rest have different owners.¹⁰

This analysis answered the first research question: there is no geolocated, complete and exhaustive inventory of urban parks. Given the disparity between the available databases, it was considered that none met the characteristics imposed in the research question. Much less was there an integral system of urban parks found, that is, some free software or urban planning tool that allows visualizing parks on an urban scale.

Statistical Capacity of the DPJ

As the ^{Corbett and Mellouli (2017)} model suggests, beyond the operational and administrative sphere where the DPJ seems to be currently located, it had to be suggested that its registry of parks and other green spaces should also include the “political” and “sustainability” dimensions, meaning, not limiting itself to the area under its responsibility, but serving the production of indicator 11.7 at the local level of the 2030 Agenda.

Officials and ex-officials of the DPJ agree that in the last 20-30 years this unit has diminished in terms of financial and human resources. A former official recalled that in the 1997-2000 triennium, when it was called “maintenance and forestry”, there were much more personnel (official 4, personal communication, January 28, 2022). Said unit then had more than 300 staff members and they were very well equipped. However, today they are half the number they were in the year 2000 and their equipment is limited. The latter, in his opinion, demonstrates the little importance that green spaces have represented for past municipal administrations.

“They would not have the capacity to take care of all the green spaces” mentioned another former official interviewed who also held a position at the management level. For this reason, “civil society was called, organized groups, around 50-60 groups. They liked to organize activities in favor of the environment, such as afforestation and cleaning of green spaces”, the idea was that they would adopt a park (official 9, personal communication, January 15, 2022). In this regard, an official on duty corroborated that “only 28-33% of the total areas are tended to per month, which means it takes three months to go around all the places” (official 3, personal communication, 9 February 2022).

On the other hand, some interviewed actors who worked in the DPJ for more than 10 to 15 years, recalled having compiled detailed inventories of squares, boulevards, and sites served by this department, however, their effort was not followed through with the change of administration. Regarding the DPJ list of parks, an interviewee stated: “I remember that the DPJ had an average of 670, but once we began to see parks where certain people spoke to us, we calculated around 240 or 260 additional parks that are not included in the DPJ list” (official 9, personal communication, January 15, 2022).

An official on duty commented that they had just purchased commercial software to work with the GIS, that they were in the process of locating all the spaces, but they had few staff, so they wished the state universities would help them. Although they would be limited only to the areas under their responsibility.

Undoubtedly, the agency that possesses the technical capabilities and skills to create a complete and exhaustive inventory of existing urban parks would be the Implan. However, the information they handle is not public (in editable format) and the E6 plan as such, would need to be recategorized, verified and specified to serve as a source of information and generate sustainability indicators. Such capabilities would undoubtedly be complemented by DPJ’s field experience.

The qualitative analysis of the interviews with the key actors who have worked on the subject helped to answer the third research question: the existing difficulties in having an official inventory of urban parks. As suggested by ^{Feltynowski et al. (2018, p. 1)}, “officials often focus on a fraction of urban green spaces. They refer only to those for whom they are formally responsible”.