Natural Endowments and Planning Interventions: The Spatio-Temporal Evolution and Policy Drivers of Urban Park Distribution in Shenzhen

Abstract

Research traditionally examines the spatial distribution of urban parks through the lens of spatial equity, overlooking the intricate interaction between the physical foundation of park construction and historical processes. Grounded in the theory of material geography, we investigate the mechanisms underlying the spatio-temporal evolution of urban parks in Shenzhen. We conduct topographical analysis and examine relevant historical policy texts to explore the ‘production of nature’ in China’s post-Mao urbanisation. We find that the distribution of urban parks in Shenzhen is not merely a result of social choice but a product of the interplay between material natural endowments—centred on topography—and urban spatial policies across historical stages. During rapid urbanisation, government-led spatial policies functionally reorganised and assigned symbolic meanings to diverse topographical features, such as plains, hills, and coastal areas, transforming them into urban parks that support capital accumulation and urban upgrading. The proposed ‘topography–policy’ synergistic framework transcends neutral spatial descriptions, revealing the nexus between the commodification of nature and urban governance. We clarify the rationale for the creation of contemporary urban green spaces in China and offer novel theoretical and empirical insights into sustainable urban transformation worldwide.

1. Introduction

Urban greening is a prominent global policy trend. The construction of urban parks—along with the urban–nature nexus they manifest—is profoundly embedded within the historical processes, topographical conditions, institutional arrangements, and economic systems of cities [1]. Research generally regards urban parks as spaces that promote social interaction, health, and recreation, emphasising the ecosystem services they offer and their functions as public amenities [2]. Such studies often focus on issues such as accessibility and equity and the impact of parks on real estate prices [3,4,5,6]. They tend to view social factors, such as class, ethnicity, gender, and institutional policies, as determinants of the spatio-temporal distribution of urban parks [7]. However, they generally overlook how topography, a critical material factor, interacts with urban spatial policies to influence the spatio-temporal distribution of these green spaces.

Shenzhen, a representative Chinese city of the post-Mao era [8], is a product of the global political–economic restructuring of the 1970s and mainland China’s reform and opening-up. It possesses a rich historical and policy context, as well as diverse topographical features, supporting the construction of numerous urban parks. Shenzhen had only three urban parks in 1980, but by 2020, the number had exceeded 1000. These parks were built during different phases of Shenzhen’s urban development, shaped by contemporary urban policies. They are distributed across various parts of the city and situated in distinct physical topographic contexts: plains, mountains, and reclaimed land. These two dimensions mutually constitute the basic spatio-temporal configuration of Shenzhen’s urban parks. Therefore, Shenzhen provides an ideal spatial arena for exploring the materiality of urban parks and the dynamic interaction between topography and urban spatial policies.

1.1. Urban–Nature and Topography Studies

Since the 1960s and 1970s, neo-Marxist geography has linked the natural environment to broader political–economic structures. Breaking with positivist epistemologies, these scholars recognise the interconnectedness of society and nature, as well as the relationship between the city and nature [9]. Within this intellectual lineage, natural elements in the city are not merely physical embellishments of urban form, imbued with romantic overtones and universal benevolence; rather, they are likewise constructed through urban processes shaped by political and economic contestation [10,11]. This also challenges discourse that posits an essential distinction between artificial and natural elements in the city, allowing the two to be discussed within the same analytical frame.

However, the pronounced anthropocentrism of neo-Marxist geography means that natural elements are often relegated to passive vehicles for the injection of urban meaning. Consequently, the influence of nature, a non-human agent, on urban form has failed to receive sufficient scholarly attention. Not until the 1990s did scholars point out that the ‘cultural turn’ in geography, emphasising meaning, representation, and ideology, should not supplant research on the materiality of natural elements [12]. Neo-Marxist geography’s excessive emphasis on the ‘production of nature’ within the capitalist system has obscured how natural entities shape the living environment. The diversity and agency of the Earth itself, including rivers, land, climate, topography, and plants and animals, have thereby faded into the background. The rise of ‘material geography’ within human geography signifies an acknowledgement of the effects generated by a diverse material world [13,14]. This is not a call to revert to a decontextualised spatial science that dichotomises society and nature; rather, it calls for a deeper understanding of the material world and an attention to how specific materialities interact with humanistic meanings [15].

Topography constitutes the most fundamental material basis of urban construction, including buildings, parks, and cities. Within the built environment, although understanding of the relationships between social factors and the physical environment requires refinement, the importance of topography as a material element must be emphasised. Materiality must be integrated with urban processes to avoid reducing the built environment to a mere simplified expression of social relations. Scholars of the built environment, such as Ian McHarg, and John Simonds [16,17], address the significance of topography, focusing on concrete physical landforms and technical methods for working with them, including resource surveys, evaluation, conservation, and planning and design. Other scholars conceptualise topography as a medium linking society and nature in the built environment [18]. Topography should be conceived not as a static landform but as the very process of its own formation: an unfolding trajectory shaped by profound interplays between social forces, political agendas, and physical topography [19]. Such interpretations frequently carry a positivist orientation and tend to overlook the reciprocal influences among society, politics, and topography. Definitions of topography in military and political geography offer a productive alternative: topography encompasses not only the physical features of the Earth’s surface but also human interactions with those features [20]. Topography is not limited to land; it also includes rivers, oceans, the atmosphere, and all forms of life [21]. These theoretical enquiries demonstrate that topography is co-constituted through its continuous interaction with humankind, rather than existing as a physical object external to society. Concurrently, empirical research indicates that topography possesses its own agency, actively shaping socio-economic dynamics such as wealth inequality [22], road network morphology [23], real estate valuations [24], and agricultural production [25].

1.2. Research on the Temporal and Spatial Distribution of Urban Parks

This study examines how urban topography intersects with urban spatial policies across different periods to shape the spatio-temporal distribution of urban parks. In the wake of the environmental movement, scholars came to recognise that urban parks are not natural spaces preserved in a pristine state but imbued with distinct social and political attributes. Not only their location, size, and form but also the timing and sites of and motives for their construction are closely tied to considerations such as land value and urban policy [26]. Nevertheless, limited research asks how material elements such as topography interact with social and political contexts to influence the spatio-temporal distribution of urban parks. Material geography reminds us that focusing exclusively on social factors risks engendering an overly anthropocentric understanding of the spatio-temporal distribution of urban parks.

People have long understood urban parks as green spaces that enhance citizens’ well-being and quality of life. However, the environmental movement fundamentally challenged this understanding [27]. Scholars increasingly recognise that neoliberal forms of urban governance have transformed the relationship between people and nature in cities, exposing urban parks to issues such as capital accumulation, urban growth, and social exclusion [28,29]. Consequently, a substantial body of critical scholarship addresses the relationships between the spatio-temporal distribution of urban parks and class, gender, and racial inequalities. These studies demonstrate that inequalities in the spatio-temporal distribution of urban parks vary across spatial scales and influence not only their area and distribution but also their internal configuration [30,31,32]. The construction of urban parks, therefore, forms part of neoliberal urban governance rather than being a straightforward act of benevolence [33]. Social factors significantly shape the broader spatio-temporal distribution of urban parks [34]. For example, after the collapse of Romania’s socialist regime, urban green spaces in Bucharest significantly changed in number and distribution [35]. Similarly, after the implementation of neoliberal development strategies, the number of urban parks in Mexico progressively declined, with their distribution becoming increasingly uneven across districts [36]. Moreover, most studies of the spatio-temporal distribution of urban parks actually focus only on their spatial distribution at a particular moment, rather than on their spatio-temporal patterns within historical urban development. Yet such patterns reveal clear historical characteristics, and only by examining them dynamically can we understand the origins of present conditions and evaluate future possibilities [37,38,39].

In sum, the inequalities emphasised by neo-Marxist geography undoubtedly merit in-depth examination, but this analytical framework privileges the dominant influence of social factors on geographical phenomena. Material geography critically reworked this framework. From the perspective of material geography, research on the spatio-temporal distribution of urban parks cannot be conducted solely around social factors; it must return to topography, the most fundamental material basis upon which urban parks are constructed. Furthermore, studies on the spatio-temporal distribution of urban parks tend to adopt a synchronic rather than diachronic approach, thereby failing to appreciate the historical dynamics underlying parks’ spatial distribution.

1.3. Urbanisation and Urban Parks in Post-Mao China

Rapid urbanisation in China is the product of profound interaction and contestation between state power and local governments [40]. This process has catalysed the emergence of cities and the precipitous expansion of urban space, which often assimilates multiple rural characteristics [41]. These rural vestiges are manifested, on the one hand, in the rapid conversion of former peasants into urban citizens, whose lifestyles and social networks nonetheless retain distinctive agrarian traits [42]. On the other hand, collective land—including cropland, orchards, fish ponds, and oyster beds—has been nationalised for urban construction, triggering profound socio-natural changes [43]. The socio-natural transformations occurring within the process of Chinese urbanisation are difficult to explain using simplistic Western urban theories [44,45]. For instance, scholars propose the concept of the ‘sustainability fix’ to describe the integration of environmental objectives into urban governance [46]. However, this process is primarily oriented towards realising economic goals; within the context of Chinese urbanisation, it fails to address the root causes of environmental challenges. Building upon this critique, the concept of the ‘socio-ecological fix’ highlights the importance of transforming socio-natural relations to achieve environmental development. Rather than being driven by urgent economic imperatives, this approach emphasises the intervention of state power to realise the environmental objectives associated with the ‘eco-state’ or ‘eco-urbanism’ [47,48].

Amidst China’s ongoing urbanisation, the development of urban parks has evolved from a mere greening initiative into a core strategy driving spatial governance. As China’s urban population has expanded over recent decades, the demand for urban park construction has risen commensurately. This trend is characterised by a higher demand for development in large than small cities [49]. During the early stages of the reform and opening-up era, special economic zones (SEZ)—exemplified by Shenzhen—undertook the large-scale construction of municipal parks. A primary objective of this endeavour was to sculpt the physical landscape of an ‘entrepreneurial city’ to attract an inflow of global capital [50]. In the 21st century, park development became further embedded within the framework of local governments’ ‘land finance’ strategies. By institutionally transforming natural terrain into parkland, governments not only ensured the rigid protection of ecological baselines but also leveraged the resulting ‘environmental premium’ to induce rapid appreciation in the value of surrounding land [51,52]. This form of landscape production—driven jointly by policy and capital—often manifests spatially as uneven development: high-quality park resources are preferentially allocated to strategic growth poles with high output potential, such as new urban districts and high-tech industrial parks. This practice has triggered a significant disparity in environmental well-being between these emerging development zones and high-density, older urban districts [53,54]. This intricate interplay between function and space renders China’s urban parks an ideal lens through which to observe the underlying logic of ‘socio-ecological restoration’ under state-led governance.

International experience likewise reveals the profound interaction between policy and terrain. Freshkills Park transformed a landfill into an ecological park through interventions including landfill capping and the restoration of salt marsh wetlands. Yet scholars have argued that this highly engineered form of “artificial nature” attempts to obscure the environmental costs of urban expansion and consumerism, thereby effacing questions of environmental injustice [55]. London introduced the “Pocket Parks” initiative, converting small and residual parcels—such as sloped sites and leftover urban spaces—into community green areas [56]. Berlin, meanwhile, adopted a strategy of “Temporary Use,” opening post-industrial vacant land to the public through low-intervention approaches [57]. Most of these Euro-American practices are embedded within frameworks of neoliberal governance, emphasising public–private partnerships, market mechanisms, and community participation. In contrast to these approaches, the construction and renewal of urban parks in China have unfolded amid rapid industrial upgrading and urbanisation, relying on top-down measures such as state land ownership, ecological control lines, and land reclamation to transform natural terrain—including mountainous and coastal landscapes—into park resources that serve capital accumulation. Shenzhen represents a paradigmatic case of this developmental trajectory.

In summary, the spatio-temporal distribution of urban parks—products of the interplay between social, political, and economic forces—is deeply embedded within the logics of state governance and capital accumulation. However, current theoretical frameworks remain inadequate for analysing this complex socio-natural transformation amidst China’s post-Mao urbanisation. First, although neo-Marxist geography illuminates the role of power in restructuring natural landscapes, it emphasises the unidirectional dominance of social factors, obscuring the agency of topography as a material substrate. Second, while discussions specific to China touch on the socio-ecological fix and the logic of land management, they fail to analyse in depth how topography, a material element, participates in the production of urban space. As an extreme case study of China’s rapid urbanisation and institutional transition, the evolution of Shenzhen’s urban park system represents far more than a mere greening initiative; rather, it constitutes a strategic process through which the government institutionally restructures differentiated topographies to realise environmental premiums and transform spatial governance. Proposing a framework of ‘topography–policy’ co-evolution, this study explores—within a dynamic, diachronic context—how topography couples with the logic of government-led socio-ecological restoration to drive the spatio-temporal evolution of urban parks. This endeavour deepens our understanding of the logic of the ‘production of nature’ in transitional China but also offers a theoretical perspective, grounded in material geography, on spatial production and ecological governance in high-density cities globally.

We focus on urban parks in Shenzhen constructed between 1980 and 2020, employing a mixed qualitative and quantitative approach to examine the topographical foundations of park construction and explore the relationships between urban topography, urban policy, and urban parks. Specifically, we address the following two questions. (1) If urban parks are understood as material landscapes constructed upon specific topographies, how do urban parks built in Shenzhen in different periods express their materiality? What topographical characteristics do they exhibit? (2) How have Shenzhen’s urban policies transformed specific topographies into urban parks? How do topography and policy interact? (3) What new characteristics are exemplified by the development of urban parks in Shenzhen, as an emerging city that rose to prominence in the post-Mao China?

2. Materials and Methods

Among the 1093 parks listed in the official inventory released by the Shenzhen municipal government at the end of 2019, we classified them into three groups according to their size, as shown in

Table 1. Total Number and Area of Urban Parks by Size Category.

Area of Individual Park	Numbers	Proportion in Terms of Number	Area (Hm)	Proportion in Terms of Area
<1 Hm	627	57.37%	257.1	0.65%
1–300 Hm	445	40.71%	7724.2	19.64%
>300 Hm	21	1.92%	31,354.6	79.71%
Sum	1093	100.00%	39,335.8	100.00%

:

According to the definition proposed by the International Federation of Parks and Recreation Administration [2] and the Chinese industrial standard of Standard for classification of urban green space [58], green spaces larger than 300 hectares are not, strictly speaking, urban parks; rather, they are more appropriately classified as country parks. Unlike urban parks, the siting of such spaces is determined almost entirely by natural conditions—such as large-scale mountainous terrain and water bodies—rather than by the implementation of urban policy. Among the 1093 parks surveyed, the 21 country parks account for as much as 79.71% of the total park area. Such an extreme distribution is not conducive to generating broadly representative conclusions, and excluding these cases helps maintain the analytical focus on the urban parks central to this study. In addition, the 627 micro-parks smaller than 1 hectare constitute only 0.65% of the total area. Since our analysis is primarily based on areal statistics, these micro-parks exert only a marginal influence on the overall results. Moreover, the development of such micro-parks often involves multiple actors and highly fragmented land ownership, while official records frequently contain ambiguous construction dates or evidence of repeated renaming and redevelopment. Because the core concern of this study is spatiotemporal evolution, inaccurate construction dates would substantially interfere with the identification of evolutionary patterns. Excluding these highly uncertain samples therefore improves the reliability of the analysis. Accordingly, this study focuses on urban parks ranging from 1 to 300 hectares as the primary objects of analysis and statistical investigation.

We constructed a ‘topography–policy’ framework to analyse these urban parks, as shown in Figure 1. In the process of Shenzhen’s urban spatial development, four years—1980, 1992, 2005, and 2012—represent key milestones (

Table 2. Key policies in the history of Shenzhen’s urban spatial development.

Year	Policies	Significance
1980	Regulations on the Guangdong Special Economic Zone	Shenzhen was established as a city.
1992	Interim Provisions on Rural Urbanisation in the Shenzhen Special Economic Zone	Land in the four administrative districts south of the Second Line was nationalised.
2005	Measures on Urbanisation Land Management in Bao’an and Longgang Districts of Shenzhen Municipality Shenzhen Basic Ecological Control Line Management Regulations Plan for the Coordinated Development of the PRD Metropolitan Area	Land in the administrative districts north of the Second Line was nationalised. Protection of important ecological spaces witthin the city limits. The integration of the Pearl River Delta was included in the government’s agenda.
2012	Overall Plan for Land-Management System Reform of Shenzhen	Shenzhen’s spatial development entered a phase of stock utilisation.

). We thereby divided the construction of Shenzhen’s urban parks into four phases, 1980–1992, 1993–2004, 2005–2012, and 2013–2020, and compiled data on the number, total area, and spatial distribution of parks built during each phase (Table 2 and

Table 3. Basic information on urban parks built in Shenzhen during the four focal periods.

	Time Period	First Period (1980–1992)			Second Period (1993–2004)
Location		Total Area	Number	Average Area	Total Area	Number	Average Area
South of Second Line		347.2	13	26.7	1030.4	30	34.3
North of Second Line		71.8	9	8.0	607.5	54	11.3
Total		419.0	22	19.0	1637.9	84	19.5
	Time Period	Third Period (2005–2012)			Fourth Period (2013–2020)
Location		Total Area	Number	Average Area	Total Area	Number	Average Area
South of Second Line		1067.9	43	24.8	531.6	60	8.9
North of Second Line		2044.3	103	19.8	1836.5	133	13.8
Total		3112.2	146	21.3	2368.1	193	12.3

and Figure 2). Using the official list of parks and OpenStreetMap (https://www.openstreetmap.org/, accessed on 22 June 2025), we obtained vector boundaries for the 445 selected parks, which were then manually corrected according to Google Earth. We further employed digital elevation model topography data from the European Space Agency’s Copernicus Open Access Hub (https://scihub.copernicus.eu/, accessed on 15 July 2025) in ArcGIS Pro to calculate the average elevation and slope for each park. Another key focus of this study was the land use origin of these 445 parks; that is, their land use status when Shenzhen was established as a city in 1980. For this, we obtained 1985 Shenzhen land use data from the CLCD dataset (https://zenodo.org/records/12779975, accessed on 02 August 2025) and manually corrected these data to 1980 using US Geological Survey satellite imagery (https://earthexplorer.usgs.gov/, accessed on 05 August 2025) to reconstruct Shenzhen land use in 1980 (this approach was feasible because between 1980 and 1985, urban construction in Shenzhen was concentrated in the Luohu and Shekou areas. Additionally, very few parks were built in these two areas, making the workload for manual corrections extremely small.). The vector boundaries of the parks were then overlaid to determine the land sources for each park.

Finally, we classified the parks into eight categories according to their area, average slope, and average elevation (

Table 4. Parameters for categorising urban parks by area, slope, and elevation.

Area (Hm)	A1	A2	A3	A4	A5	A6	A7	A8
	1–30	30–60	60–90	90–120	120–150	150–180	180–210	>210
Average Slope (°)	S1	S2	S3	S4	S5	S6	S7	S8
	0–3	3–6	6–9	9–12	12–15	15–18	18–21	>21
Average Elevation (m)	E1	E2	E3	E4	E5	E6	E7	E8
	0–15	15–30	30–45	45–60	60–75	75–90	90–105	>105

). By combining these classifications with the land sources of parks in each period, we summarised the topographical characteristics of parks constructed in the four phases and analysed the factors contributing to these topographical patterns across periods.

3. Results

3.1. Area and Number

As shown in Figure 3 and Figure 4, across the four periods, the third period exhibited the largest total park area (3112.2 hectares) and the fourth period had the largest number of parks (193). The total area of urban parks constructed in the city increased from the first period (419.0 hectares) to the third period (3112.2 hectares), before declining slightly in the fourth period (2368.1 hectares). In contrast, the total number of parks rose steadily from the first period (22 parks) to the fourth period (193 parks), resulting in a noticeable decrease in the average area per park during the fourth period. Figure 4 further shows that in the fourth period, parks within the A1 category accounted for the largest proportion of total park area (29.9%); the proportion for all other categories remained below 20%. Additionally, the third and fourth periods both featured large parks in the A7 and A8 categories. In the third period, the combined area of these large parks (991.5 hectares) was substantially greater than that in the fourth period (598.0 hectares), with no parks of this size constructed during the first or second period.

Before 2010, the Shenzhen Special Economic Zone (SSEZ) encompassed only the area south of the Second Line; after 2010, it was extended to cover the entire city. Before this expansion, however, infrastructure development policies differed between the areas north and south of the Second Line, influencing the construction of urban parks. During the first and second periods, parks south of the Second Line accounted for a greater total area than those to the north, whereas in the third and fourth periods, the total area of parks constructed north of the Second Line exceeded that of parks constructed to the south (Figure 5). In terms of number, with the exception of the first period, when there were 13 parks south of the Second Line and 9 to the north, the number of parks built north of the Second Line in all periods was almost twice that of parks built to the south (Figure 6).

From a policy perspective, the first period corresponds to the early stage of Shenzhen’s establishment, when the policy environment of reform and opening-up was relatively unstable and remained exploratory. Urban development at this stage was therefore limited in scale, and urban parks were correspondingly small in size, consisting primarily of parks in the A1–A2 categories. In the second period, the reform and opening-up policy was reaffirmed at the national level. The Interim Provisions on Rural Urbanisation in the Shenzhen Special Economic Zone stipulated the full nationalisation of land south of the Second Line, yet mounting land constraints compelled urban expansion northward across this boundary. As a result, both urban development and park construction expanded significantly, and parks in the A5–A6 categories began to emerge. In the third period, the measures on urbanisation and land management in Bao’an and Longgang Districts of Shenzhen Municipality extended land nationalisation to areas north of the second line. In addition, the implementation of the Shenzhen Basic Ecological Control Line Management Regulations enabled large mountainous areas to be formally incorporated into the urban park system. This institutional change led to the emergence of larger parks in the A7–A8 categories. In the fourth stage, the implementation of the Overall Plan for Land-Management System Reform of Shenzhen further intensified citywide land scarcity. Under these conditions, urban park development shifted again toward smaller-scale projects, with A1–A3 category parks becoming dominant. From a policy perspective, the first period corresponds to the early stage of Shenzhen’s establishment, when the policy environment of reform and opening-up was relatively unstable and remained exploratory. Urban development at this stage was therefore limited in scale, and urban parks were correspondingly small in size, consisting primarily of parks in the A1–A2 categories. In the second period, the reform and opening-up policy was reaffirmed at the national level. The Interim Provisions on Rural Urbanisation in the Shenzhen Special Economic Zone stipulated the full nationalisation of land south of the Second Line, yet mounting land constraints compelled urban expansion northward across this boundary. As a result, both urban development and park construction expanded significantly, and parks in the A5–A6 categories began to emerge. In the third period, the measures on urbanisation and land management in Bao’an and Longgang districts of Shenzhen Municipality extended land nationalisation to areas north of the second line. In addition, the implementation of the Shenzhen Basic Ecological Control Line Management Regulations enabled large mountainous areas to be formally incorporated into the urban park system. This institutional change led to the emergence of larger parks in the A7–A8 categories. In the fourth stage, the implementation of the Overall Plan for Land-Management System Reform of Shenzhen further intensified citywide land scarcity. Under these conditions, urban park development shifted again toward smaller-scale projects, with A1–A3 category parks becoming dominant.

3.2. Slope and Elevation

In terms of slope, during the first period, urban park construction primarily occurred on relatively gentle topography (Figure 7). In the second and third periods, parks were more likely to be built on steeper slopes. In the fourth period, however, numerous parks were once again built on gentler slopes. Figure 8 shows that most of the parks built during the first period were in the S3 category, making up 55.3% of the total park area during that time. Almost no parks were built on slopes above the S5 category. By the second and third periods, park construction had gradually expanded to steeper slopes, ranging from S5 to S8. In the fourth period, however, a significant number of parks were built on gentler slopes, specifically in the S2 (30.0%), S3 (17.1%), and S4 (13.4%) categories, while the proportion of park area in the S5–S8 categories was markedly lower than in the second and third periods.

The trends in elevation closely mirrored those observed for slope. During the first period, urban parks were primarily constructed on lower-elevation topography (Figure 9). In the second and third periods, a larger number of parks were built in higher-elevation areas, whereas in the fourth period, some parks were again established on lower-elevation topography. As shown in Figure 10, in the first period, park construction was largely concentrated in the E1 (45.2%) and E2 (37.7%) categories, with very few parks located in E3 or higher categories. In the second period, the combined proportion of parks in the E3–E6 categories reached 71.6%, while the combined proportion in the E1 and E2 categories declined significantly, and almost no parks were in the highest categories, E7 and E8. By the third period, however, some parks had been constructed in the higher-elevation E7 (11.6%) and E8 (15.6%) categories, and the proportion of parks on lower-elevation topography (E1, E2, and E3) had noticeably decreased compared with the previous two periods. In the fourth period, the proportion of parks in the mid-elevation categories, E4 and E5, declined relative to the third period, while some parks were again established on lower-elevation topography in E1, E2, and E3.

The slope and elevation of urban parks are also closely related to urban policy. In the first period, both urban development and park construction took place within the Luohu Plain south of the Second Line in Shenzhen. Accordingly, urban parks were primarily characterised by slopes in the S2–S3 category and elevations in the E1–E2 category. In the second period, urban park development began to expand northward into the hilly and mountainous terrain beyond the second line. This shift led to the emergence of parks with slopes in the S5–S7 category and a substantial number of parks in the E6 category in terms of elevation. In addition, land scarcity south of the second line also pushed park construction into relatively steeper mountainous areas such as Lianhua Mountain and Bijia Mountain. In the third period, after the full nationalisation of land north of the second line, a large number of urban parks appeared with slopes in the S5–S8 category and elevations in the E4–E8 category. In the fourth period, under the policy framework of Pearl River Delta regional integration, many large-scale urban parks were developed in the coastal areas along the eastern shore of the Pearl River Estuary. This resulted in a renewed concentration of parks with slopes in the S2–S4 category and elevations in the E1–E3 category.

3.3. Land Use

In terms of land use, urban parks in Shenzhen were constructed primarily on farmland (34.5%) and forestland (54.0%), the predominant forms of land existing at the time of the city’s establishment in 1980. As shown in Figure 11 and Figure 12, the trend of constructing parks on farmland continued to increase in terms of area, but after the second period, this growth became only slight. In contrast, the area of parks constructed on forestland grew rapidly during the first three periods but declined substantially in the fourth period (1111.8 hectares) compared with the third period (2026.5 hectares). The share of parks constructed on farmland was as high as 63.2% in the first period, gradually declining to 25.7% in the third period, before recovering to 37.6% in the fourth period. The proportion of parks built on forestland, in contrast, increased steadily from 22.8% in the first period to 65.1% in the third period, before falling to 47.0% in the fourth period. Additionally, the use of water bodies for urban park construction in Shenzhen is noteworthy. In the first period, parks occupied only 35.7 hectares of water, but because the total park area was relatively small (419.0 hectares), this represented 8.5% of the park construction area. During the second (116.7 hectares, 7.1%) and third (253.9 hectares, 8.2%) periods, the area of parks occupying water bodies increased, but the proportion remained similar to that in the first period. By the fourth period, however, this proportion had risen to 14.5%, with the area of parks constructed on water bodies increasing to 342.5 hectares.

The topographical changes and trends of urban parks constructed across the four periods were generally as shown in

Table 5. Characteristics of Shenzhen urban parks: Area, count, slope, elevation, and land use by construction period.

Time period	First Period (1980–1992)	Second Period (1993–2004)	Third Period (2005–2012)	Fourth Period (2013–2020)
Area and Number	The total area (347.2 hectares) was limited, and the number of parks (13) was small. The constructed parks were primarily in the A1 and A2 categories. The number of parks built south of the Second Line (13) exceeded that of parks built north of it (9).	Both the total area (1030.4 hectares) and the number of parks (30) increased. The constructed parks were mainly in the A1, A2, A4, and A5 categories. The number of parks built north of the Second Line (54) exceeded that of parks built south of it (30).	The total area increased slightly (1067.9 hectares), and the number of parks continued to rise (43). Large parks (in the A7 and A8 categories) emerged, accounting for a relatively large proportion. The number of parks built north of the Second Line (103) exceeded that of parks built south of it (43).	The total area dropped markedly (531.6 hectares), but the number of parks rose further (60). The proportion of A7 and A8 category parks decreased overall compared with the preceding period, while the proportion of A1 and A2 category parks increased. More parks were built north of the Second Line (133) than south of it (60).
Slope and Elevation	Urban parks were primarily built in areas with gentle slopes (S3, S4, S5) and low elevations (E1, E2).	Urban parks were gradually built in areas with moderate slopes (S2, S3, S5, S6) and low-to-medium elevations (E1, E3, E4, E6).	Urban parks continued to be developed in areas with moderately high to high slopes (S5, S6, S7, S8) and mid-to-high elevations (E4, E5, E6, E7, E8), and parks began to appear in areas characterised by steep slopes and high elevations.	The proportion of parks constructed in areas with steep slopes and high elevations declined compared with the previous period, whereas the share of parks developed in areas with gentle slopes (S2, S3, S4) and low elevations (E1, E2, E3) increased significantly.
Land Use	Farmland constituted the primary source of land for park construction (63.2%), followed by forestland (22.8%).	The proportion of parks constructed on farmland declined (39.4%), whereas the proportion constructed on forestland increased (51.2%).	The proportion of parks constructed on farmland continued to decline (25.7%), while the proportion constructed on forestland continued to rise (65.1%).	The proportion of parks constructed on water areas (14.5%) increased significantly compared with the first three periods (7.1–8.5%).
Policy Drivers	With the implementation of the Regulations on the Guangdong Special Economic Zone, Shenzhen was officially established as a city. During this period, the policy framework of reform and opening-up remained relatively unstable, and urban development was limited in scale. Construction activities were primarily concentrated on the agricultural plains south of the Second Line, as designated by the government, resulting in substantial conversion and occupation of cultivated land.	Due to increasing scarcity of land resources south of the Second Line, urban expansion extended into the hilly areas north of the Second Line, resulting in increased occupation of forest land and a decline in the use of cultivated land. In addition, as the policy of reform and opening-up was further consolidated and the Interim Provisions on Rural Urbanisation in the Shenzhen Special Economic Zone were implemented, urban development continued to expand, and the average size of individual urban parks increased.	Urban expansion continued into the hilly areas north of the Second Line. With the implementation of the Shenzhen Basic Ecological Control Line Management Regulations, large mountainous areas were incorporated into the urban park system. Concurrently, the process of Pearl River Delta regional integration began, leading to extensive urban development and land reclamation along the eastern shore of the Pearl River Estuary.	Land resource constraints emerged on both sides of the Second Line. With the implementation of the Overall Plan for Land-Management System Reform of Shenzhen, the Pearl River Delta regional integration process entered a critical stage, and a large number of urban parks began to be developed on reclaimed land.

. In terms of both area and number, the peak period of urban park construction in Shenzhen occurred during the third period (2005–2012), with the focus of development gradually shifting from the area south of the Second Line, which dominated the early years of city construction, to the north. It was during this third period that Shenzhen began to establish large parks in the A7 and A8 size categories, most of which were located north of the Second Line, including Longcheng Park, Dayun Park, Pingluanshan Park, and Pinghu Ecological Park. From the first to the third period, the average area per park showed a slight upward trend, but the result declined markedly in the fourth period, reaching only 12.3 hectares. Regarding slope and elevation, parks in the first period were generally constructed on gentle slopes (S1–S3) and lower-elevation topography, such as Honghu Park, Litchi Park, and People’s Park in central Luohu. In the second and third periods, parks tended to be built on moderately steep to steep slopes and medium-to-high elevation areas, with parks established even on steep and high-elevation topography (S7–S8, E7–E8), such as Meilin Park, Xiaonanshan Park, and Qiushui Mountain Park. This trend did not continue into the fourth period, during which numerous parks were constructed on gentler and lower-elevation topography, such as Talent Park and Xiwan Mangrove Park. In terms of land use, farmland and forestland at the time of the city’s establishment constituted the primary sources of land for park construction. In the first period, parks were mainly built on farmland. In the second and third periods, the proportion of parks built on both farmland and forestland increased, with forestland showing the most notable rise. In the fourth period, the proportion of parks constructed on forestland declined, whereas the use of water bodies as a source of land for park construction became more prevalent compared with the previous three periods.

From a land use perspective, the first period of urban park development was concentrated in the Shenzhen Luohu area, which, prior to reform and opening-up, consisted largely of agricultural land. As a result, parks constructed on former farmland accounted for the highest proportion during this period. In the second and third periods, the implementation of land nationalisation policies and the Basic Ecological Control Line policy meant that, both south and north of the Second Line, urban park development was predominantly located in mountainous and hilly terrain. These areas correspond to the primary distribution of forest land, making forest land the dominant land-use type for urban parks constructed in these two periods. In the fourth period, some urban parks were also developed on former cropland and forest land. Notably, however, there was a marked increase in parks constructed on areas previously classified as water bodies. This shift is closely associated with the Pearl River Delta regional integration strategy, which redirected the spatial focus of urban park development toward the eastern shore of the Pearl River Estuary.

4. Urban Parks’ Topographical Evolution and Their Policy Contexts

Superficially, the siting of urban parks may appear to result from multiple seemingly random decisions. However, the results above indicate that parks constructed in Shenzhen during different periods have exhibited clear trends and regularities in terms of their topographical characteristics. To better understand these patterns, we offer an explanation from the perspective of urban spatial development, as follows.

4.1. From Plains to Mountains: Insufficient Space for Development

At the time of its establishment, the SSEZ was designated by the State Council to cover 327.5 square kilometres. Excluding steep mountainous areas deemed unsuitable for development, the planned usable area was 160 square kilometres, of which 123 square kilometres were intended for construction and development [59]. By 1989, however, 109 square kilometres of land within the SEZ had already been allocated, leaving very little of the originally planned development area remaining [60].

The scarcity of spatial resources within the SEZ compelled the government to consider a ‘Full Territorial Development’ strategy, pushing urban expansion into hilly and mountainous areas north of the Second Line. The primary challenge was land tenure. Under the Land Administration Law of the People’s Republic of China, urban construction, including residential areas, factories, and urban parks, can only occur on state-owned land; farmers’ land is collectively owned and cannot be directly used for urban development. Before 2004, land north of the Second Line in Shenzhen was predominantly rural collective land, and the government could only acquire land use rights for urban construction through a ‘one plot at a time’ approach, constraining the implementation of Full Territorial Development [61]. In June 2004, the Shenzhen municipal government introduced measures for the management of urbanised land in the Bao’an and Longgang districts of Shenzhen to fix this problem. These measures quickly converted 956 square kilometres of collective land outside the Second Line into state-owned land through administrative means. Moreover, under this policy, forestland and land on steep slopes (exceeding 25 degrees), which had previously been excluded from construction, were also directly converted into state-owned land [62], enabling urban development at the land tenure level. Following the removal of these property rights barriers, both the area and number of urban parks outside the Second Line increased substantially after the third period. The total area of urban parks expanded rapidly from 607.5 hectares in the second period to 2044.3 hectares in the third period, while the total number also increased substantially from 54 to 103 parks, as shown in Figure 5.

Although the hilly and mountainous areas north of the Second Line were converted into state-owned land, if only land tenure were considered, without regard for the integrity of ecological space, these areas could theoretically be used for urban construction. In the Shenzhen City Master Plan (1996–2010), urban space north of the Second Line was located on numerous small plains among the hills and mountains, forming urban clusters interconnected by roads to create a network structure; the surrounding hills and mountains functioned as ecological buffers between the clusters [63]. Rapid urbanisation north of the Second Line in the mid-1990s severely undermined this planning concept. Much of the land designated for construction breached the boundaries of the urban clusters, encroaching on the ecological buffer zones in the hilly and mountainous topography, and the clusters gradually merged into contiguous urban forms. Consequently, the government had to affirm the integrity of ecological spaces. In November 2005, the Shenzhen Basic Ecological Control Line Management Regulations were implemented. The control line encompasses the following [64]:

• Primary water source protection areas, scenic spots, nature reserves, contiguous high-quality farmland protection zones, forests, and country parks.
• Mountainous areas and forestland with slopes exceeding 25%, as well as highland with elevation surpassing 50 metres within the Special Economic Zone and 80 metres outside the SEZ.
• Major rivers, reservoirs, and wetlands.
• Environmental corridors and green spaces essential for maintaining ecosystem integrity.
• Islands and coastal land areas with significant ecological conservation value.
• Other regions that require basic ecological control.

These regulations stipulate that the 974 square kilometres of land in the ecological control zone cannot be used for general urban construction, such as building factories or residential housing. However, the regulations also specify the following permissible construction activities within this zone:

• Major road and transportation facilities
• Municipal public utilities
• Tourism facilities
• Urban parks
• Facilities related to agriculture, education, and scientific research that are compatible with ecological and environmental protection.

Paradoxically, while the ecological control line effectively curtailed most forms of urban expansion, urban parks were positioned as a privileged exception within this regulatory framework. The area within the ecological control line accounts for nearly half of Shenzhen’s total land area. Within the remaining 1000-plus square kilometres available for development, at least 270 square kilometres were designated for industrial use, with the rest of the land needed to accommodate various urban functions, including residential areas, commercial and service facilities, transportation infrastructure, and public amenities. On the one hand, the government needed to provide relatively flat and developable land for these various urban functions; on the other hand, it had to construct urban parks to meet the recreational needs of a rapidly growing population. Siting parks in steep and high-elevation areas during the third period represented a compromise that addressed both competing demands, as there were no restrictions on urban park construction within the ecological control line. Compared with the second period, the area of urban parks located in S6–S8 category increased by 14.0% during the third period, while the area in E7–E8 category increased by 13.3%. By contrast, the area of parks in S1–S3 category decreased by 10.0%, and the area in E1–E3 category declined by 17.0%, as shown in Figure 8 and Figure 10.

In 1989, the Shenzhen Construction Bureau, in collaboration with the China Academy of Urban Planning and Design, jointly formulated the Urban Development Strategy of Shenzhen. Serving as the preliminary research report for the preparation of the Shenzhen City Master Plan (1996–2010), this development strategy provided a detailed assessment of the relationship between topography and urban development in Shenzhen (

Table 6. Assessment of city-wide topography in Urban Development Strategy of Shenzhen (1989).

	Western	Central	Eastern
Favourable factors	Good conditions of land use, flat terrain, free from natural topographic threshold	Relatively concentrated development, prone to aggregated economy effect	Many types of shoreline, conducive to comprehensive development
Adverse factors	Relatively dispersed structure of the city, not conducive to the organisation of urban life	Many mountainous areas, difficult traffic organisation, and high construction costs of new transport facilities	Many mountainous areas and high development difficulty

):

As the most fundamental material condition for urban construction, topography interacts with the formulation of urban policies through mutual adaptation. This synergy shapes the fundamental spatial configuration of the city and, accordingly, determines the underlying patterns of urban park distribution. The topographical conditions for urban construction in Shenzhen have differed between periods. First, this reflects the city’s enormous demand for spatial resources, while flat and developable land remains extremely limited. Second, apart from the first period, when urban construction in the alluvial plains of Luohu District faced relatively few topographical constraints, subsequent periods of development had to adapt to specific topographical conditions, such as hills, mountains, and reclaimed land, to achieve construction objectives. Urban park construction followed the same principles, resulting in distinct topographical characteristics in the spatio-temporal distribution of Shenzhen’s urban parks across periods.

4.2. From Farmland and Forestland to Sea: Changes in Land Use

Although urban parks differ from grey infrastructure such as factories and residential areas due to their verdant appearance, they share the same political and economic context. In 1980s Shenzhen, global processes of spatial production, coupled with local urban development, transformed the agricultural landscape, activating the latent economic value of land and producing a modern urban landscape [65]. Farmland and forestland from the planned economy era were rapidly converted for more economically productive uses, including factories, residential areas, and urban parks, all of which served the broader process of urbanisation.

At the beginning of the reform and opening-up in 1983, farmland and forestland accounted for 27.55% and 53.47% of Shenzhen’s total land area, respectively [66]. To accommodate the transfer of international manufacturing, Shenzhen constructed many industrial facilities and associated residential areas. This rapid urbanisation led to the extensive occupation of contiguous farmland in the early 1980s. By the end of 2019, the area of farmland in Shenzhen had declined to only 2844.74 hectares, representing 1.44% of the city’s total land area. Although forestland also decreased significantly between 1983 and 2019, in the latter year it still accounted for 32.5% of the city’s total land area [67]. Forested areas in Shenzhen are spatially aligned primarily with hilly and mountainous topography [68], which is less conducive than farmland to rapid urban development. As a result, despite rapid urbanisation, a substantial area of forestland remained in 2019. It is thus unsurprising that Shenzhen’s urban development was largely built upon pre-reform farmland. However, more detailed insights into urban parks—particularly their characteristics and trends across the four periods—are needed.

During the first period, many of Hong Kong manufacturing enterprises relocated to the SSEZ. At that time, urban development—including factories and residential areas—was concentrated in Luohu, Shekou, and Shatoujiao [69]. Before the city’s establishment, these flat, coastal alluvial plains were predominantly farmland, with small areas of tidal flats [62]. When urban parks were constructed on such topography, farmland served as the primary land source for their development.

During the second and third periods, Shenzhen rapidly advanced its Full Territorial Development strategy, extending urban construction north of the Second Line. Consequently, the conversion of forestland for urban park construction began to increase significantly. In these northern areas, forestland was predominant, while farmland was limited. As a result, the area of urban parks constructed on farmland showed only marginal growth after the second period, and its proportional share even declined from 39.4% in the second period to 25.7% in the third period, as shown in Figure 11. In the mid- to late 1980s, the municipal government assessed the city’s land development potential. Forestland was identified as the land use category with the largest remaining capacity for development (

Table 7. Assessment of the city’s land development potential (1989).

	Suitable Development Area (km2)	Utilised Area (km2)	Remaining Development Capacity (km2)
Paddy field	425.77	268.19	157.57
Dry land	133.13	96.56	36.56
Forestland	1119.99	541.84	578.16
Pasture	48.55	5.40	43.15
Aquaculture water	83.50	42.15	41.35
Tidal flats	40.22	10.21	30.01

) [60].

Despite the considerable remaining development potential of forested areas, the overwhelming demand for land hindered the government’s ability to execute timely conservation initiatives in mountainous and hilly areas. Unregulated construction accounted for about 50% of the urban area north of the Second Line [70]. Throughout this period, mountains and hills linked to forested areas formed the topographical basis for extensive urban development, including urban parks. Meanwhile, south of the Second Line, where land resources were similarly constrained, urban parks were increasingly sited on hilly topography that was otherwise difficult to develop, such as Lianhua Mountain Park, Bijia Mountain Park, and Meilin Park. This explains why forestland became the primary source of land for urban park construction during the second period. In the third period, the Shenzhen municipal government issued Regulations on the Management of the Basic Ecological Control Line, which strictly limited general urban construction, such as residential and industrial development, on hills and mountains. Urban parks, however, were exempt from these restrictions. This further encouraged the use of forest land as the primary land source for urban park construction, resulting in an increase in the area of parks built on forest land from 837.9 hectares in the third period, with the proportion rising from 51.2% to 65.1% (Figure 12 and Table 5).

In the fourth period, the focus of urban construction shifted toward the eastern bank of the Pearl River Estuary and the Shenzhen Bay area. Consequently, there was a marked decrease in urban parks with forestland with mountainous and hilly topography as their primary land source. In contrast, the proportion of urban park areas created by encroaching on water bodies increased significantly. Meanwhile, owing to proximity to the coastal plain, the proportion of parks located on gentle slopes and low-elevation topography also rose compared with the previous period. In terms of slope, the area of urban parks located in the S1-S3 category increased by 19.8%, while the area of parks situated in the E1–E3 category increased by 18.8% (Figure 8 and Figure 10). In the 1989 Urban Development Strategy of Shenzhen, land reclamation in the western areas near the Pearl River Estuary had already been considered [60]:

The city possesses a coastline of 229.96 kilometres. The western section belongs to the plain coast type, featuring a straight shoreline… Furthermore, due to the rapid siltation process, land reclamation is relatively easy to implement…

By 2009, during the formulation of the Shenzhen City Master Plan (2010–2020), extensive land reclamation operations were already underway in this region. In this version of the master plan, the western coastline was designated a ‘strategic development area’: an area with immense growth potential whose optimal development timing had yet to mature, including the Qianhai District and the Shajing Western Riverside District:

The Qianhai District possesses significant locational advantages… with a favourable coastal shoreline, providing the necessary conditions for establishing a potential Central Business District (CBD). The plan positions it as an integral part of Shenzhen’s 21st-century core urban area and one of the city’s two future primary centres. The Shajing Western Riverside Area is required to serve as a spatial reserve for the construction of world-class high-end industrial parks.

This period witnessed the successive completion of numerous coastal parks on reclaimed land, such as the OCT Wetland Park, Shenzhen Bay Park, Talent Park, and Xiwan Mangrove Wetland Park. The establishment of the Qianhai–Houhai Center and New Marine City on the eastern bank of the Pearl River Estuary directly follows the boundaries of these parks. Concurrently, urban spaces previously occupied by low-end manufacturing underwent urban renewal to align with Shenzhen’s strategic role as a core city in the Pearl River Delta regional integration. From the perspective of spatial policy, the construction objectives of these urban parks are highly consistent with the surrounding urban development; both are intrinsically linked to attracting high-end industrial talent, promoting industrial upgrading, and facilitating regional integration [71].

Topography strongly shapes land use patterns, which also determine the physical conditions of land available for urban park construction. Urban parks built in Shenzhen during different periods also exhibit corresponding trends in their land sources. The reasons for such trends are twofold. First, Shenzhen originated from Bao’an County, which was dominated by agriculture, and nearly all urban development, including urban parks, was carried out on the former farmland, forestland, and tidal flats of Bao’an. Second, urban development in different periods occurred across areas with distinct topographical conditions, and compared with other types of urban construction, urban parks were less constrained by topography. As a result, the land sources of urban parks constructed in different periods display distinct spatio-temporal characteristics.

4.3. From City to Region: Land Reclamation and Regional Integration

The logic of Shenzhen’s urban spatial development is not rooted solely within the city but fundamentally regionally based, particularly in the 21st century. Consequently, urban parks have evolved into a spatial extension of regional development strategies. Since the 1980s, cities in the Pearl River Delta have developed on an export-oriented economic basis, which means that other cities, apart from Guangzhou and Shenzhen, are not strongly dependent on the service functions of their own urban centres. After China joined the World Trade Organization, Shenzhen no longer enjoyed exclusive preferential policies, prompting Hong Kong and international investors to focus on a broader regional context. In turn, under the framework of integrated regional development, competition among cities in the Pearl River Delta intensified, leading each to place greater emphasis on the service functions of its urban centre to prevent high-end industries from relocating to neighbouring cities. In response, cities across the Pearl River Delta converged on development strategies aimed at strengthening the service capacity of their own urban centres [72]. For Shenzhen, establishing a coastal urban centre in the west, along the eastern shore of the Pearl River Estuary, allowed the city to both integrate into the Pearl River Delta and enhance the service functions of its centre. However, by this time, land in areas such as Shajing and Fuyong along the estuary had already been occupied by low-end manufacturing and associated facilities that had relocated in the 1990s. Consequently, Shenzhen was compelled to employ land reclamation to accommodate its new urban centres and foster the growth of high-end service industries and high-tech sectors. Through this approach, Shenzhen sought to reduce its dependence on Hong Kong and international manufacturing transfers, progressively develop local industries, foster independent innovation, and secure a more central position within the global urban network [73].

In the process of constructing an urban centre, urban parks foster an atmosphere of innovation, particularly when the city centre shifts to coastal areas, as seafront parks provide a stronger visual impact and aesthetic environment than other types of parks. This aligns with Shenzhen’s goal of developing a high-quality urban centre to reinforce its position as a central city within the Pearl River Delta. A new urban centre oriented towards high-end services and high-tech industries must cater to the spatial imaginaries of dominant technological and managerial elites, encompassing residential and recreational spaces, concentrated urban functions, and proximity to arts, cultural, and entertainment hubs [74]. Within the discourse of ecological aesthetics, urban parks serve as spatial media expressing elite environmental aesthetics and lifestyles. These parks aestheticise urban space through specific forms of natural elements, not only attracting tech and management elites to live and work in the vicinity but also serving as a material medium for global corporate headquarters to project their international image and influence [75,76,77]. This aligns perfectly with Shenzhen’s strategic requirement to strengthen its role as a core city in the Pearl River Delta by developing high-quality urban centres. This objective is exemplified by the Shenzhen Nanshan District Territorial Spatial Plan (2021–2035), which, in its coastal spatial planning, designates Shenzhen Bay as a demonstration zone for headquarters economy, high-end culture, and coastal tourism and leisure [78].

From the perspective of Shenzhen’s urban spatial development, the integration of the Pearl River Delta has driven the westward shift of urban central areas, including large urban parks (Figure 13). During the first period, the SEZ attracted investment from Hong Kong and international sources, and urban service functions were concentrated around the old town of Luohu. Litchi Park, People’s Park, Honghu Park, and Donghu Park were all constructed in Luohu old town or on adjacent coastal alluvial plains. As spatial saturation occurred in Luohu centre, the Shenzhen City Master Plan (1996–2010) proposed relocating the new city centre westward to Futian District, initiating the second and third periods of urban park development. During this phase, central parks such as Shenzhen Central Park, Bijia Mountain Park, Huanggang Park, and Meilin Park were successively established on the hilly topography of Futian. Subsequently, during the fourth period, the Shenzhen City Master Plan (2010–2020) formalised a dual-centre spatial layout of Qianhai–Houhai and Futian–Luohu, leading to the construction of parks on reclaimed land in the west, including Shenzhen Bay Park, Talent Park, and Coastal Cultural Park. By the time the Shenzhen Territorial Spatial Master Plan (2020–2035) designated the Shekou Peninsula and Shenzhen Bay in Nanshan District the city’s sole ‘comprehensive innovation core area’, Nanshan was already a hub of highly visible parks, earning the reputation of a ‘district of a hundred parks’ [79]. Parks built on reclaimed land not only provide recreational space for citizens, as conventional urban parks do, but also leverage the aesthetic value of sea views to attract high-tech and managerial elites, thereby promoting urban industrial upgrading [71].

Land reclamation is a key issue in Shenzhen’s urban spatial development. This method of acquiring urban space represents the interaction between local topography and the localisation of Pearl River Delta regional integration policies. Over several decades, Shenzhen’s city centre has gradually shifted westward, shaping the spatio-temporal distribution of urban parks south of the Second Line. Newly constructed urban parks have been progressively aligned with topography, whether plains, hills and mountains, or coastal areas. On the one hand, at the macro level, their siting and internal facilities have been structurally shaped by regional integration policies and urban spatial planning. On the other hand, at the material level, they have been designed and built upon different types of topography, resulting in urban parks with diverse spatial forms and layouts.

5. Discussion

5.1. Government-Led Urbanisation and the Construction of Urban Parks

In Shenzhen, the construction of urban parks is not merely intended to provide public services; rather, it constitutes a form of ‘production of nature’, embedding topography, a natural element, within the city’s socio-political and economic fabric. Throughout this process, the government functions not only as a regulator of space but also as a creator of market conditions [80]. Through the dual land nationalisation policies of 1992 and 2004, the government authoritatively reconstructed the significance of diverse topographical features, including plains, hills, and water bodies. These features were transformed from mere natural endowments into production factors that could be integrated into the system of capital accumulation. This essentially represented a spatial strategy implemented by the government to facilitate a comprehensive transition towards a market economy. Consequently, the development of urban parks has become a strategic instrument for the government to generate assets and drive urban upgrading. Such governmental intervention in nature exemplifies how, since the reform and opening-up era, Chinese municipal governments have utilised institutional frameworks to transmute natural topography into productive factors. This ‘commodification of topography’ can be understood as a refinement of the Marxist geographical concept of the ‘commodification of nature’. Through spatial policy instruments, the state redefines plains, mountainous areas, and coastal zones as differentiated and scarce resources that can be mobilised to serve capital accumulation.

An analysis of Chinese urban space necessitates attention to the power and role of the state and the government [81]. Whilst capital is significant, its efficacy is contingent upon the recognition and endorsement of state power through various policy mechanisms. This logic fundamentally underpins the production of nature and space within urban parks.

5.2. Regionalisation, Industrial Upgrading, and Environmental Equity

In the fourth stage, the primary focus of urban park construction in Shenzhen exhibited a conspicuous shift towards the eastern shore of the Pearl River Estuary. This represents not merely a geospatial relocation but a quintessential manifestation of ecological gentrification [82]. During this period, urban parks were predominantly established within high-tech industrial belts and elite residential enclaves for management professionals in districts such as Nanshan and Bao’an. In essence, these parks evolved into landscapes designed to serve high-end industrial talent, creating a feedback loop where environmental aesthetics, advanced technology, and ecological gentrification reinforce each other [83]. This locational logic reveals a distinct characteristic of environmental equity in Shenzhen: through the commodification of the scarce coastal topography, the state has curated a series of iconic, aestheticised spaces designed to attract globalised enterprises and elite human capital. The commodification of topography, under the post-Mao urban governance regime in which economic development and industrial upgrading constitute primary political objectives, also functions as an expression of politicisation.

This spatial orientation of green space, which privileges high-end industrial zones, stands in stark contrast with traditional old urban areas, which are densely populated and space-constrained. In these older districts, environmental improvement often remains stagnant, as land resources were exhausted during earlier waves of industrialisation and now offer limited potential for capital premiums. Such differential allocation illustrates how urban park construction has been deployed as a form of spatial fix. Rather than prioritising the recreational needs of the public, it primarily converts natural endowments into stratified welfare to bolster industrial upgrading. Under these circumstances, coastal parks in Shenzhen, particularly in Nanshan and Bao’an, are not entirely public natural phenomena; instead, they function to some extent as exclusionary symbols of gentrification, thereby exacerbating geographical asymmetries in the distribution of environmental well-being.

5.3. Research Limitations and Future Research

In this study, we primarily focus on elevation, slope, and land use, while omitting other topographic indicators, such as aspect and ruggedness. We also do not systematically examine socio-economic variables such as population, income, and educational attainment in the surrounding areas of urban parks. Future research will incorporate these two categories of variables and quantitatively assess their correlations as well as underlying causal relationships. Furthermore, future research could develop more actionable policy recommendations based on such quantitative assessments, moving beyond historical analysis.

6. Conclusions

Following the environmental movements of the 20th century, the analytical paradigm of neo-Marxist geography reintroduced nature to the urban context, highlighting the inequities generated by capitalism’s methods of ‘producing nature’. This knowledge paradigm continues to exert a profound influence, prompting mainstream research to focus on social dimensions such as urban power and class, to a certain extent veering toward social determinism and thereby obscuring the agency of urban parks as material entities.

Inspired by material geography, this study finds that the distribution of urban parks in Shenzhen is not merely the outcome of social choice but rather the product of the interaction between material nature endowments—centred on topography—and urban spatial policies across different historical periods. This finding moves beyond the conventional emphasis on spatial equity that has dominated previous research.

The ‘topography–policy’ framework proposed here essentially functions as a material record of the ‘production of nature’ within Shenzhen’s urbanisation process. As a material factor, topography is far from being a neutral geographical backdrop; instead, it constitutes a contested site of power struggles orchestrated by urban policy. This trajectory—shifting from the early construction of urban parks on plains and hills to the subsequent production of maritime vistas on reclaimed land—clearly demonstrates how topography has been transformed into a system of capital accumulation. From a Marxist perspective, nature within the urban fabric evolves into a direct material extension of the logic of economic development. Furthermore, urban parks, as material entities, actively participate in the actualisation of policy objectives. This dynamic interplay between topography and policy represents not only the embedding of social meaning but also the mutual shaping of power and the physical world. The spatio-temporal distribution of urban parks in Shenzhen can thus be interpreted as a political history of topography, offering a material–geographical lens through which to critically examine the sustainability models of high-density cities.

Our contribution to the literature lies in, first, decoupling urban parks from the academic context of the ‘urban–nature dichotomy’. We argue that the spatial distribution of urban parks is not a random formation but significantly governed by urban policies, guided by the political economic logic of industrial development and capital accumulation. This is particularly true of Shenzhen, a city built in the post-Mao era, where the development of urban parks can, in essence, be understood as the state-led politicisation and commodification of urban topography. Second, proceeding from topography, the most critical physical foundation for the development of urban parks, we examine how urban policies across different periods have interacted with urban topography, thereby endowing urban parks in various periods with distinct topographical characteristics. Furthermore, topography plays an active role in driving the upgrading of urban industries. Finally, this research serves as a reminder that sustainable urban development, along with its associated green infrastructure initiatives, must not be conceived solely in two-dimensional plans; rather, it requires a rigorous assessment of the potential and dynamic agency inherent in the undulating urban topography. Although this study focuses on Shenzhen, the ‘topography–policy’ framework offers valuable insights for other high-density cities. Future research could further leverage big data methodologies to explore how such topographical variations directly influence citizens’ micro-level perceptions and the equity of their usage.