A Multi-Criteria Framework for Sustainable Marine Spatial Planning in Coastal Cities: Case Study in Shenzhen, China

Abstract

As rapid urbanization intensified pressure on coastal ecosystems, balancing economic development with ecological preservation remained a critical challenge. This study developed a multi-criteria framework for integrated marine spatial planning and applied it to Shenzhen, China—a rapidly expanding coastal metropolis overseeing 1145 km² of marine territory with a 260.5 km coastline, 61.47% of which has been anthropogenically modified. The methodology combined ecological, environmental, and socioeconomic indicators through a hierarchical evaluation system, using entropy-weighted indices and GIS-based spatial analysis to assess marine space suitability across three functional categories: ecological protection, urban development, and biological resource utilization. The results revealed that 38.53% of Shenzhen’s coastline retains natural geomorphology, while 23.7% annual growth in maritime infrastructure projects since 2015 highlights urgent development pressures. Marine spatial zoning identified priority areas for ecological conservation, urban–industrial expansion, and biological resource utilization through a 1 km × 1 km grid-based analysis, integrating water quality monitoring data. The framework demonstrated how adaptive zoning of underutilized coastal areas could enhance resource efficiency while balancing economic and environmental goals. These findings provide empirical evidence for optimizing marine resource allocation in coastal megacities, emphasizing the importance of data-driven planning anchored in quantitative metrics (e.g., shoreline development intensity and fisheries resource carrying indices) to achieve long-term sustainability.

1. Introduction

Marine zoning in coastal areas is essential for achieving a balance between protecting marine ecosystems and fostering economic growth [1]. As urbanization and industrialization along coastlines continue to accelerate, the importance of establishing clear and effective marine zoning has become increasingly evident [2,3,4]. This type of zoning facilitates the efficient utilization of marine resources, safeguards biodiversity, and promotes sustainable development [5]. By defining specific areas for ecological protection, resource extraction, and industrial development, marine zoning aims to prevent overexploitation and environmental degradation, while also fostering the growth of marine industries in a way that is both responsible and beneficial to the local economy [6]. In addition, effective zoning can address challenges such as the fragmentation of land–sea planning and the need for integrated management of coastal and marine environments [7]. This comprehensive approach is essential for achieving long-term ecological, economic, and social sustainability in coastal regions [8,9].

From an international perspective, the rise of marine spatial planning marks a shift in marine governance from single-resource management to systemic spatial coordination [10,11,12]. For instance, in 2014, the European Union enacted the Marine Spatial Planning Directive, which mandated that member countries establish marine functional zones by 2021, aiming to address the competing interests of shipping, fishing, energy development, and environmental conservation [13]. This framework not only emphasizes spatial optimization of economic activities but also guides zoning decisions by quantifying ecosystem service values (such as carbon sequestration capacity and biodiversity maintenance). For instance, in the North Sea region, compatibility zones have been designated for wind farms and bird migration corridors [14,15]. Transboundary cooperation has become another important trend, with Baltic Sea countries establishing unified data-sharing platforms and joint management mechanisms to coordinate fisheries quotas and pollution control standards [16,17]. Mediterranean coastal countries, through regional marine agreements, have incorporated tourism development and coral reef protection into zoning plans, reflecting the collaborative nature of global marine governance [18,19].

The current research hotspots focus on adaptive zoning to address climate change, the application of multi-objective optimization models, and the improvement of stakeholder participation mechanisms [20,21]. With the ongoing rise in global sea levels and the increasing frequency of extreme weather events, conventional static zoning approaches are inadequate for addressing the dynamic needs of risk management [22,23]. For instance, in Shanghai, a coastal buffer zone has been designated in the eastern part of Chongming Island, restricting permanent building development and reserving flexible space for tidal flooding [24]; Guangzhou’s Nansha New Area has adjusted high-risk zones to ecological wetland parks based on simulated storm surge inundation ranges, enhancing the city’s resilience [25]. Technological innovations have further promoted the scientific advancement of zoning. Spatial analysis models based on GIS and remote sensing technologies can integrate multidimensional data, such as hydrological, biological, and economic data [26]. For example, in Zhejiang Province, machine learning algorithms are used to predict the relationship between farming density and red tide occurrence in the Xiangshan Port area, optimizing the spatial layout of fishing zones [27]. At the same time, community participation mechanisms are increasingly becoming an important part of zoning decisions. Hainan Province established an ecological compensation fund for enterprises in the development of Sanya Bay, directing a portion of tourism revenues to coral reef restoration, reflecting the co-governance logic of multiple stakeholders [28].

However, marine zoning in coastal cities still faces several challenges. First, the inadequacy of data support and assessment systems restricts the refinement of zoning plans. Many regions lack high-precision marine environmental data [29]. Secondly, land resource scarcity exacerbates the conflict between protection and development. For example, in the Beibu Gulf of Guangxi, the overdevelopment of the Yintan Scenic Area has resulted in beach erosion and water quality deterioration, forcing the government to invest hundreds of millions of yuan in ecological restoration [30]. Moreover, the rigid framework of traditional zoning planning is inadequate to adapt to sudden environmental changes [31]. Local governments’ GDP evaluation mechanisms still prioritize the layout of ports, industries, and other economic functions, while the regulatory enforcement of ecological protection areas remains insufficient [32].

This study addressed the central research question of how a multi-criteria framework could integrate ecological, environmental, and socioeconomic indicators to achieve sustainable marine spatial planning in rapidly urbanizing coastal cities, with a focus on balancing economic development and ecological preservation. The investigation was carried out through the development and application of a hierarchical evaluation system in Shenzhen, China, which centered on three primary objectives. First, it aimed to identify priority zones for the conservation of natural coastlines, which made up 38.53% of Shenzhen’s coastline, while mitigating the pressures exerted by human activities, as 61.47% of the coastline had been modified. Second, the study tackled the challenge of meeting urgent infrastructure needs, given the 23.7% annual growth in maritime projects since 2015, while minimizing negative impacts on the environment. Lastly, it sought to optimize the utilization of biological resources, such as fisheries and tourism, while adhering to environmental limits like water quality and sediment standards. The research tested the effectiveness of a data-driven framework, which combined entropy-weighted indices, GIS analysis, and adaptive zoning to address the conflicts between these priorities in coastal megacities like Shenzhen.

This paper emphasized the significance of adopting an integrated approach to marine planning to foster discussions around sustainable development. It examined the intricate interactions between coastal ecosystems, human populations, and environmental shifts, offering valuable insights into the necessity for collaborative management practices. The study provided practical recommendations to enhance the resilience of both marine ecosystems and coastal communities, highlighting the importance of sustainable resource utilization and advocating for increased cooperation among governments, local stakeholders, and researchers. In conclusion, it reinforced the critical role of integrated planning in ensuring the long-term sustainability of marine environments.

2. Methodology

The methodology employed a systematic approach to evaluate the suitability of marine spatial zoning. Initially, the marine zones were categorized into three primary functional spaces: ecological protection, urban construction, and biological resource utilization, each further subdivided into specific functional areas (e.g., ecological control, industrial use, fisheries) (in Section 2.1). The suitability evaluation integrated ecological, environmental, and socioeconomic factors through a hierarchical indicator system (in Section 2.2).

This study tested three hypotheses derived from the challenges outlined in the Shenzhen case. The first hypothesis (H₁) posits that a multi-criteria framework integrating ecological, environmental, and socioeconomic indicators can systematically resolve conflicts between coastal development and conservation. The second hypothesis (H₂) suggests that entropy-weighted indices and GIS-based spatial analysis can reduce subjective bias in zoning decisions compared to traditional expert-driven methods. The third hypothesis (H₃) asserts that the framework is replicable in other coastal cities facing similar urbanization pressures, as long as local data and policy contexts are properly integrated.

2.1. Marine Zone Division System

The assessment of marine spatial zoning suitability was founded on the development feasibility evaluations of three key functions: ecological protection, agricultural production, and urban construction. Tailored to the unique characteristics of marine environments, the evaluation involved processes such as basic data processing, indicator selection and weighting, and the calculation and classification of evaluation units. Ultimately, this led to the determination of the suitability assessment results for marine ecological spaces, construction use spaces, and marine biological resource utilization spaces within the jurisdictional waters. Furthermore, by integrating zoning determination standards, a clear marine ecological zoning scheme was established.

Building on this foundation, and in conjunction with the original marine functional zoning, the “Land and Sea Classification Guidelines for Territorial Spatial Investigation, Planning, and Use Regulation (Trial)” and the current regional conditions and development needs, a small-scale marine spatial zoning system was defined [33,34] (

Table 1. Small-scale marine spatial zoning system.

Types of Marine Space	Function Areas
Marine Ecological Space	Ecological Protection Areas
	Ecological Control Areas
Construction Marine Space	Urban Reserved Areas
	Industrial Communication Use Areas
	Transportation Use Areas
Marine Biological Resource Utilization Space	Fisheries Use Areas
	Tourism and Recreation Use Areas

). This study proposed a specific layout for the marine functional zoning of the coastal city, which primarily comprised marine ecological spaces, marine biological resource utilization spaces, and construction use spaces.

Marine ecological spaces were defined as marine areas characterized by natural attributes, primarily serving the functions of providing ecological services, protection, and ecosystem maintenance. Construction use spaces referred to marine areas designated for activities related to the development of ports, coastal industries, and major infrastructure projects. Marine biological resource utilization spaces were identified as marine zones with good environmental quality and high primary productivity, suitable for supplying marine products, marine biomedical raw materials, and marine tourism.

2.2. Marine Development Suitability Evaluation

This paper approached the topic from the perspective of marine spatial classification and regional conditions [5,35,36,37,38,39,40,41]. Based on the distinct attributes and characteristics of each space type, it developed a suitability evaluation indicator system, drawing upon a secondary indicator classification, as presented in Table 2.

The establishment of the zoning suitability evaluation indicator system is theoretically grounded in internationally recognized principles of marine spatial planning and China’s specific policy contexts. The system aligns with the EU Maritime Spatial Planning Directive’s three pillars through operationalized metrics: Ecological coherence is quantified via biodiversity indices like the Shannon–Wiener index and habitat quality assessments; ecosystem-based management is implemented through ecological carrying capacity indicators, while blue growth objectives are measured by development intensity and transportation accessibility parameters [42,43]. This hierarchical structure follows UNESCO’s pressure–state–response framework, where shoreline development intensity represents anthropogenic pressures, water quality compliance rates and sediment standards reflect environmental states, and ecological control zoning serves as the management response [44].

Indicator selection adhered to three validation criteria: spatial explicitness ensured 1 km² mappability for coastal management applications, policy relevance guaranteed alignment with China’s Marine Functional Zoning GB/T 17108-2006 standards, and scientific validity incorporated peer-reviewed indices like the ecological health index (EH_indx) [45].

Table 2. Zoning suitability evaluation indicator system.

Evaluation Objective	Secondary Indicator	Tertiary Indicator	Indicator Definitions
Marine Ecological Space	Marine Ecosystem	Ecological Importance	Comprehensive assessment of the importance of protected marine areas, typical marine habitat protection, and marine topography protection.
		Marine Biodiversity	Calculated using Shannon–Wiener index $H^{*}=-\sum _{i=1}^s{p}_i{log}_2{p}_i$, $\mathrm{where} p_i$ is the proportion of individuals of species i, and s is the total number of species [46,47].
		Ecological Health Status	Calculated using ecological health status ${EH}_{indx}={\sum }_i^p{INDX}_i$, where ${INDX}_i$ represents the health status of indicator i, and p is the number of indicator categories [48].
	Marine Ecological Environment Evaluation	Marine Environmental Carrying Index	Ratio of the area of marine regions meeting water quality standards to the total area.
		Marine Ecological Carrying Index	Based on the “Guidelines for Coastal Marine Ecological Health Evaluation” [49,50,51].
	Marine Environmental Quality	Seawater Quality	According to “Seawater Quality Standards” [52].
		Marine Sediment Quality	According to “Marine Sediment Quality Standards” [53,54,55].
		Marine Biological Quality	According to “Marine Biological Quality Standards” [56,57].
Construction Marine Space	Marine Space Resource Evaluation	Shoreline Development Intensity	Calculated using shoreline development intensity $P_A={\displaystyle \frac{l_{mB}\times q_B+l_{mT}\times q_T{+l}_{mG}\times q_G+l_{mH}\times q_H}{l_{total}}}$, where ${ l}_{total}$ is the total shoreline length. $l_{mB}$$, l_{mT}$$, l_{mG}$ and $l_{mH}$ represent the lengths of the coastal lines of the enclosure dam, protective embankment, industrial and urban areas, and port and dock areas, respectively. $q_B$, $q_T$, $q_G$, and $q_H$ represent the impact levels of the four types of artificial coastlines on the marine resource environment [58].
		Marine Area Development Intensity	Calculated using marine area development intensity $P_E={\displaystyle \frac{\sum _{i=1}^n\left(S_i\times l_i\right)}{S}}$, where $S_i$ is the area of type i, S is the total area of marine functional zones, and $l_i$ is the resource consumption coefficient for type i [59].
	Resource Environment Constraints	Marine Development Potential	Evaluated through the minimum of shoreline $D_s$ and tidal flat resource utilization potential $D_t$. $D_s=1-{\displaystyle \frac{L_r+L_p}{L_z}}$, where $L_r$ is the length of developed coastline, $L_p$ is the length of coastline within marine protected areas, and $L_z$ is the total length of coastline. $D_t=1-{\displaystyle \frac{S_r+S_p}{S_z}}$, where $S_r$ is the area of developed intertidal zone $S_p$ is the area of intertidal zone within marine protected areas, and $S_z$ is the total area of intertidal zone [60].
	Land Reclamation Resource	--	Calculated as reclaimed land area minus area confirmed for reclamation projects.
	Transportation Advantage	Distance to Residential Areas Distance to Ports Distance to Roads Distance to Offshore Islands	The Delphi method is used to determine the importance of transportation factors. The assigned values are as follows: distance to residential areas is 0.1, to roads 0.25, to ports 0.35, and to offshore islands 0.3. To find the distance of transportation factors from the evaluation unit’s center, extreme value normalization is used to calculate the relative distances [61,62,63,64].
Marine Biological Resource Utilization Space	Marine Fisheries Resource Evaluation	Fisheries Resource Carrying Index	Fisheries resource carrying index $F=F_1\times 0.6+F_2\times 0.4$, where $F_1$ is the swimming animal index and $F_2$ is the fish egg and larval fish index. $F_1={\displaystyle \frac{ES+TL}{2}}$, where ES is the proportion of economically important species in the catch, and TL is the average trophic level index for the coastal area. $F_2=F_E\times 0.2+F_L\times 0.8$, where $F_E$ is the fish egg density, and $F_L$ is the larval fish density [65].
	Marine Environmental Quality	Seawater Quality	Same as above.
		Marine Sediment Quality	Same as above.
		Marine Biological Quality	Same as above.
	Marine Ecosystem	Ecological Importance	Same as above.
		Marine Biodiversity	Same as above.
		Ecological Health Status	Same as above.
	Transportation Advantage	Distance to Residential Areas Distance to Ports Distance to Roads Distance to Offshore Islands	Same as above.

Table 2. Zoning suitability evaluation indicator system.

Evaluation Objective	Secondary Indicator	Tertiary Indicator	Indicator Definitions
Marine Ecological Space	Marine Ecosystem	Ecological Importance	Comprehensive assessment of the importance of protected marine areas, typical marine habitat protection, and marine topography protection.
		Marine Biodiversity	Calculated using Shannon–Wiener index $H^{*}=-\sum _{i=1}^s{p}_i{log}_2{p}_i$, $\mathrm{where} p_i$ is the proportion of individuals of species i, and s is the total number of species [46,47].
		Ecological Health Status	Calculated using ecological health status ${EH}_{indx}={\sum }_i^p{INDX}_i$, where ${INDX}_i$ represents the health status of indicator i, and p is the number of indicator categories [48].
	Marine Ecological Environment Evaluation	Marine Environmental Carrying Index	Ratio of the area of marine regions meeting water quality standards to the total area.
		Marine Ecological Carrying Index	Based on the “Guidelines for Coastal Marine Ecological Health Evaluation” [49,50,51].
	Marine Environmental Quality	Seawater Quality	According to “Seawater Quality Standards” [52].
		Marine Sediment Quality	According to “Marine Sediment Quality Standards” [53,54,55].
		Marine Biological Quality	According to “Marine Biological Quality Standards” [56,57].
Construction Marine Space	Marine Space Resource Evaluation	Shoreline Development Intensity	Calculated using shoreline development intensity $P_A={\displaystyle \frac{l_{mB}\times q_B+l_{mT}\times q_T{+l}_{mG}\times q_G+l_{mH}\times q_H}{l_{total}}}$, where ${ l}_{total}$ is the total shoreline length. $l_{mB}$$, l_{mT}$$, l_{mG}$ and $l_{mH}$ represent the lengths of the coastal lines of the enclosure dam, protective embankment, industrial and urban areas, and port and dock areas, respectively. $q_B$, $q_T$, $q_G$, and $q_H$ represent the impact levels of the four types of artificial coastlines on the marine resource environment [58].
		Marine Area Development Intensity	Calculated using marine area development intensity $P_E={\displaystyle \frac{\sum _{i=1}^n\left(S_i\times l_i\right)}{S}}$, where $S_i$ is the area of type i, S is the total area of marine functional zones, and $l_i$ is the resource consumption coefficient for type i [59].
	Resource Environment Constraints	Marine Development Potential	Evaluated through the minimum of shoreline $D_s$ and tidal flat resource utilization potential $D_t$. $D_s=1-{\displaystyle \frac{L_r+L_p}{L_z}}$, where $L_r$ is the length of developed coastline, $L_p$ is the length of coastline within marine protected areas, and $L_z$ is the total length of coastline. $D_t=1-{\displaystyle \frac{S_r+S_p}{S_z}}$, where $S_r$ is the area of developed intertidal zone $S_p$ is the area of intertidal zone within marine protected areas, and $S_z$ is the total area of intertidal zone [60].
	Land Reclamation Resource	--	Calculated as reclaimed land area minus area confirmed for reclamation projects.
	Transportation Advantage	Distance to Residential Areas Distance to Ports Distance to Roads Distance to Offshore Islands	The Delphi method is used to determine the importance of transportation factors. The assigned values are as follows: distance to residential areas is 0.1, to roads 0.25, to ports 0.35, and to offshore islands 0.3. To find the distance of transportation factors from the evaluation unit’s center, extreme value normalization is used to calculate the relative distances [61,62,63,64].
Marine Biological Resource Utilization Space	Marine Fisheries Resource Evaluation	Fisheries Resource Carrying Index	Fisheries resource carrying index $F=F_1\times 0.6+F_2\times 0.4$, where $F_1$ is the swimming animal index and $F_2$ is the fish egg and larval fish index. $F_1={\displaystyle \frac{ES+TL}{2}}$, where ES is the proportion of economically important species in the catch, and TL is the average trophic level index for the coastal area. $F_2=F_E\times 0.2+F_L\times 0.8$, where $F_E$ is the fish egg density, and $F_L$ is the larval fish density [65].
	Marine Environmental Quality	Seawater Quality	Same as above.
		Marine Sediment Quality	Same as above.
		Marine Biological Quality	Same as above.
	Marine Ecosystem	Ecological Importance	Same as above.
		Marine Biodiversity	Same as above.
		Ecological Health Status	Same as above.
	Transportation Advantage	Distance to Residential Areas Distance to Ports Distance to Roads Distance to Offshore Islands	Same as above.

This paper employed the comprehensive index method based on entropy weights to determine the weight factors for each indicator. The entropy weight method was specifically adopted to resolve three critical challenges: addressing compensation effects between ecological preservation and development priorities, modeling non-linear marine carrying capacity relationships, and objectively prioritizing zoning conflicts through information entropy calculations. This approach maintains methodological rigor while preserving the original computational integrity of the weights and system architecture, ensuring the framework’s capacity to operationalize sustainable coastal management principles without data modification or theoretical compromises. The entropy weight method is intended to determine the weights of indicators according to the information contributed by the observed values of different indicators, enabling an impartial evaluation of their relative significance [66,67,68]. The specific calculation steps for the entropy weight method were as follows:

Construct an i × j matrix using the indicator data from each region as column vectors.

Calculate the characteristic weight $P_{ij}$ of the i-th measurement for the j-th indicator using Formula (1):

$$P_{ij}={\displaystyle \frac{x_{ij}}{{\sum }_{i=1}^n{x}_{ij}}}$$

(1)

Derive the entropy for each indicator $e_j$ based on the characteristic weights using Formula (2):

$$e_j=-{\displaystyle \frac{1}{ln n}}{\sum }_{i=1}^n{P}_{ij}{ln P}_{ij}$$

(2)

Finally, determine the weights for each indicator $w_j$ using Formula (3):

$$w_j=(1-{\theta }_{ij})/{\sum }_{j=1}^n(1-{\theta }_{ij})$$

(3)

The evaluation unit acted as the primary spatial unit for analyzing the appropriateness of spatial development, leading to the establishment of an appropriate grid size that corresponded with the planning area. This study primarily utilized the dominant factor determination method, supplemented by the overlay and dynamic grid methods. Considering data availability, the marine area was divided into grid units of 1 km × 1 km. Data were collected for layer analysis and computation based on the suitability evaluation indicators for marine spatial zoning.

Subsequently, the graphical and attribute data for each indicator layer were normalized to generate raster data for every evaluation factor layer. The score for each evaluation unit was represented by the value at the grid’s central point. The extreme value method and other techniques were used for standardization, as described in the following Formula (4):

$$B_i={\displaystyle \frac{A_i-{minA}_i}{{maxA}_i-{minA}_i}}$$

(4)

where $B_i$ denotes the evaluation value of the i-th indicator; $A_i$ represents the actual value, while ${maxA}_i$ and ${minA}_i$ indicate the upper and lower limits of the actual values.

The standardized scores of each evaluation unit’s indicators and the corresponding raster data were weighted and summed through overlay operations to obtain the comprehensive score and evaluation result layer for each evaluation unit.

Finally, the comprehensive evaluation index was calculated using a multi-factor synthesis determination method [69]. Formula (5) was used:

$$Y_i={\sum }_{k=1}^n{W}_k\times Y_{ik}$$

(5)

where $Y_i$ is the comprehensive score of the i-th evaluation unit; $W_k$ represents the weight of the k-th indicator; $Y_{ik}$ is the normalized value of the k-th indicator for the i-th evaluation unit; and n is the number of evaluation units.

By utilizing ArcGIS, the natural breaks classification technique was employed to categorize the scores according to their distribution patterns. The classification results were determined according to the scores of each evaluation unit, clustering them into three levels: “high”, “medium”, and “low”. This process divided the marine space with continuous attribute values into several regions with uniform attributes, thereby yielding the suitability evaluation results. The boundaries of the merged regions were smoothed and organized, and the areas classified as “high”, “medium”, and “low” suitability were extracted as vector data, resulting in various suitability evaluation maps.

Based on the analysis of the results from the marine spatial zoning suitability evaluation, a preliminary zoning scheme was developed according to specific criteria, as presented in

Table 3. Marine spatial zoning plan based on suitability evaluation.

Suitability of Marine Ecological Space	Construction Marine Space	Marine Biological Resource Utilization Space	Type of Suggestions
High	High/Medium/Low	High/Medium/Low	Marine Ecological Space
Medium	High/Medium/Low	High	Marine Biological Resource Utilization Space
Medium	High	Medium	Construction Marine Space
Medium	Medium	Medium	Marine Ecological Space
Medium	Low	Medium	Marine Ecological Space
Medium	High	Low	Construction Marine Space
Medium	Medium	Low	Marine Ecological Space
Medium	Low	Low	Marine Ecological Space
Low	High/Medium/Low	High	Marine Biological Resource Utilization Space
Low	High	Medium	Construction Marine Space
Low	Medium	Medium	Marine Biological Resource Utilization Space
Low	Low	Medium	Marine Biological Resource Utilization Space
Low	High	Low	Construction Marine Space
Low	Medium	Low	Construction Marine Space
Low	Low	Low	Marine Ecological Space

.

In the evaluation results, only one area was identified with a high suitability level, which was designated as that specific type of space. For areas where two suitability indicators were rated as high, if one of them indicated a high suitability for marine ecological space, those areas were generally classified under marine ecological space based on the principle of prioritizing ecological protection. Otherwise, they could be prioritized for designation as marine biological resource utilization space.

In cases where all three suitability indicators were rated high, the areas were classified as marine ecological space, following the principle of ecological protection. For the areas not yet designated, if only one indicator of marine ecological suitability or marine biological resource utilization suitability was rated as moderate, those areas were classified according to that specific type of space.

For areas with two moderate suitability indicators, if the corresponding space type aligned with the main functional orientation was rated low, the designation typically followed a hierarchical order of marine ecological space, marine biological resource utilization space, and construction-related marine space. Alternatively, this could be determined based on the principle of spatial concentration. If not, the classification adhered to the principle of aligning with the main functional orientation.

When all three suitability indicators received a moderate rating, the areas were classified based on their primary functional orientation. Conversely, regions that exhibited low functional suitability overall were typically categorized as marine ecological spaces, adhering to the principle of prioritizing ecological conservation.

Given the small-scale approach of the marine spatial suitability evaluation, which centered on the integrity of spatial or ecosystem units, a localized assessment was performed, leading to certain spatial constraints. As a result, the marine spatial zoning outcomes necessitated further validation. Adjustments were made to these zoning results by referencing higher-level planning documents, such as provincial land spatial planning and coastal zone planning, to prevent conflicts.

This methodology presents three important innovations in marine spatial planning. The first is the use of an entropy-weighted multi-criteria decision framework, which contrasts with traditional subjective weighting methods. By using entropy to quantify ecological, environmental, and socioeconomic trade-offs, this approach reduces human bias and ensures a more objective balance between conservation and development priorities, with indicator weights derived from data variability. The second innovation involves a dynamic grid-based spatial analysis, which uses a high-resolution GIS model with a 1 km² grid. This model incorporates real-time environmental data, such as water quality and sediment information, along with machine learning algorithms to predict habitat vulnerability. By offering a granular approach, it addresses the limitations of static zoning, allowing for adaptive adjustments in response to coastal changes like sea level rise and infrastructure growth. Finally, the study introduces stakeholder-driven adaptive governance by incorporating participatory mechanisms, such as the Delphi method for transportation factor weighting, and climate-resilient buffer zones. This approach effectively combines technical analysis with governance, ensuring that zoning plans remain flexible and responsive to both ecological changes and community needs.

3. Case Study

3.1. Overview of the Study Area and Data

As China’s leading coastal economic hub, with unparalleled access to deep-sea basins, Shenzhen oversees a marine territory of 1145 km², boasting a 260.5 km coastline divided into distinct western and eastern segments (Figure 1 and Figure 2). The western coastline (22°32′ N–22°40′ N) stretches from the Dongbao River estuary in Bao’an District to the Shenzhen River estuary in Futian District, while the eastern coastline (22°33′ N–22°41′ N) spans from Shatoujiao in Yantian District to Bagang in Dapeng District. Notably, 61.47% (160.1 km) of this coastline has been altered by human activity, in contrast to 38.53% (100.4 km) that retains its natural geomorphological features. This spatial arrangement positions Shenzhen as a prime example of coastal megacities experiencing rapid urban growth, where escalating human-induced pressures—such as land reclamation, industrial development, and port expansion—intersect with critical marine biodiversity areas. The city’s coastal interface has thus become a critical battleground for reconciling the demands of the blue economy with the need for ecological preservation, especially considering the 23.7% annual growth rate in maritime infrastructure projects since 2015. Shenzhen’s marine environment faces numerous challenges, particularly the tension between economic activities and ecological protection amidst urban expansion and infrastructure development. As such, a comprehensive analysis of Shenzhen’s marine functional zoning offers valuable insights and lessons for other coastal cities, providing both theoretical support and empirical foundations for advancing marine management and fostering sustainable development at a higher level.

Coastline and marine data were sourced from the Shenzhen Municipal Bureau of Planning and Natural Resources, while environmental and ecological data on the marine environment were provided by the National Meteorological Center, the State Oceanic Administration, and Sentinel-2 satellite data. A set of sampling points was established to assess the current zoning of Shenzhen’s marine areas, at the following coordinates: Nan’ao (114°28′49″ E, 22°32′48″ N), Qixing Bay (114°33′6″ E, 22°33′22″ N), Shenzhen Bay (113°57′5″ E, 22°29′15″ N), and West Bay (113°50′0″ E, 22°35′36″ N). These sampling points, together with Shenzhen’s water quality monitoring stations, cover a range of nearshore marine areas, as depicted in Figure 1. The water quality parameters monitored include chlorophyll, algae, total organic carbon (TOC), chemical oxygen demand (COD), nitrogen, and phosphorus. The data collected at these sites consist of daily values, spanning from 1 January 2023 to 31 October 2024. The data used in the study are presented in

Table 4. Summary of data sources.

Data Type	Year	Format	Resolution	Source
Coastline and Marine Boundaries	2023	Shapefile	1:10,000	Shenzhen Municipal Bureau of Planning and Natural Resources
Water Quality Parameters	2023–2024	CSV	Daily, 1 km grid	National Marine Environmental Monitoring Centers (Shenzhen Bay, West Bay)
Sentinel-2 Satellite Imagery	2023	GeoTIFF	10 m (multispectral)	European Space Agency (ESA)
Marine Biodiversity Indices	2023	Raster/CSV	1 km grid	State Oceanic Administration
Socioeconomic Data (Ports, Roads)	2023	Shapefile	1:50,000	Shenzhen Urban Planning Database
Sediment Quality Metrics	2023	CSV	1 km grid	National Marine Environmental Monitoring Centers
Land Reclamation Records	2015–2023	Excel	Project-level	Shenzhen Coastal Zone Management Authority

.

3.2. Results of Marine Zone Division and Development Suitability Evaluation

The evaluation of the marine ecological environment primarily assessed the carrying capacity of the evaluation units by integrating various indicators. The evaluation of marine ecosystems focused on assessing ecological significance, emphasizing the regulatory roles of natural ecosystems and taking into account factors such as marine biodiversity and the overall health of the ecological environment. This comprehensive evaluation aimed to capture the overall status of marine ecosystems.

The results of the selected indicator weights for the marine evaluation in Shenzhen are presented in

Table 5. Weighting results of indicators used for marine evaluation in Shenzhen.

Evaluation Objective	Secondary Indicator	Tertiary Indicator	Indicator Weights
Marine Ecological Space	Marine Ecosystem	Ecological Importance	0.1958
		Marine Biodiversity	0.1005
		Ecological Health Status	0.1060
	Marine Ecological Environment Evaluation	Marine Environmental Carrying Index	0.1680
		Marine Ecological Carrying Index	0.1539
	Marine Environmental Quality	Seawater Quality	0.1075
		Marine Sediment Quality	0.0843
		Marine Biological Quality	0.0840
Construction Marine Space	Marine Space Resource Evaluation	Shoreline Development Intensity	0.1784
		Marine Area Development Intensity	0.1702
	Resource Environment Constraints	Marine Development Potential	0.1632
	Land Reclamation Resource	--	0.1521
	Transportation Advantage	Distance to Residential Areas	0.0603
		Distance to Ports	0.0906
		Distance to Roads	0.0976
		Distance to Offshore Islands	0.0876
Marine Biological Resource Utilization Space	Marine Fisheries Resource Evaluation	Fisheries Resource Carrying Index	0.1557
	Marine Environmental Quality	Seawater Quality	0.1395
		Marine Sediment Quality	0.0943
		Marine Biological Quality	0.0990
	Marine Ecosystem	Ecological Importance	0.1384
		Marine Biodiversity	0.0661
		Ecological Health Status	0.0851
	Transportation Advantage	Distance to Residential Areas	0.0554
		Distance to Ports	0.0501
		Distance to Roads	0.0721
		Distance to Offshore Islands	0.0443

.

The quality of the marine environment served as a direct reflection of the current conditions of the maritime area, with indicators such as water quality being utilized for a comprehensive assessment. The marine ecological spatial suitability evaluation results were derived by integrating assessments from these three aspects, as shown in Figure 3.

The suitability assessment of marine construction space combined evaluations of marine spatial resources, environmental constraints, and the extent of the demand for land reclamation. The analysis primarily focused on factors such as the spatial distance between evaluation units and existing reclamation areas, the spatial proximity of residential areas to marine industrial clusters, access to port facilities, and connections to major transportation routes. It also excluded high-slope coastal areas, which were identified as geomorphological restrictions that are unsuitable for reclamation or marine construction.

A comprehensive evaluation of marine construction space suitability was conducted across the entire maritime area, as depicted in Figure 3. Additionally, the evaluation process revealed that certain existing marine areas designated for construction were not particularly suitable for such purposes compared to other marine functional zoning. For regions within the current marine functional zoning that are small and possess limited potential for future expansion, it was suggested that these areas should gradually be phased out from marine construction use.

The evaluation of marine biological resource utilization space suitability was conducted by integrating various aspects, including marine fishery resources, marine environmental quality, marine ecosystems, and maritime traffic advantages. This comprehensive assessment aimed to determine the suitability of spaces for the utilization of marine biological resources, as illustrated in Figure 4. After delineating the three-zone scheme, a specific layout for marine functional zoning was proposed in accordance with the marine spatial zoning framework. Based on the evaluations of marine ecological space suitability, construction space suitability, and marine biological resource utilization space suitability, combined with established marine spatial determination standards, the marine spatial zoning plan for Shenzhen was developed. This plan formed the basis for the delineation of the marine functional zoning layout, as illustrated in Figure 4 and Figure 5.

The framework effectively balanced the competing demands in Shenzhen. Ecological preservation was given priority, with 38.53% of the coastline (100.4 km) maintaining its natural geomorphological features. These areas ranked highest in terms of ecological value and biodiversity. Urban development was strategically focused on areas with lower ecological sensitivity, such as industrial zones near existing ports like Shenzhen Bay, where indices for shoreline development intensity suggested they were suitable for expansion. Fisheries and tourism were designated for areas meeting water quality standards and possessing high transportation accessibility.

The entropy-weighted approach addressed the trade-offs between various indicators, giving precedence to ecological carrying capacity and seawater quality in the zoning process, consistent with Shenzhen’s commitment to ecosystem health. GIS analysis highlighted underutilized areas, particularly along the modified western coastline of Shenzhen, facilitating the reallocation of 34% of marine areas for urban development, while 26% was reserved for ecological preservation.

The replicability of the framework was evaluated based on its methodological transparency and adherence to international standards. The 1 km² grid system is scalable and compatible with global satellite data such as Sentinel-2 imagery. Additionally, the framework aligns with China’s Territorial Spatial Planning and the EU Marine Spatial Planning Directive, enhancing its applicability in regions with similar governance systems. Shenzhen’s approach to cross-departmental data sharing also provides a model for collaborative stakeholder engagement.

4. Discussion

The findings of this study emphasize the effectiveness of integrating ecological, environmental, and socioeconomic factors into marine spatial zoning through a systematic suitability evaluation framework. By utilizing a hierarchical indicator system and an entropy-weighted comprehensive index method, this approach addresses key limitations in traditional zoning models, particularly their static nature and limited adaptability to dynamic environmental changes. The inclusion of high-resolution spatial data and machine learning algorithms, as demonstrated in cases like Zhejiang’s Xiangshan Port, showcases the potential of advanced technologies to enhance the precision of suitability assessments. For example, the use of Shannon–Wiener biodiversity indices and ecological health metrics facilitates a quantifiable evaluation of ecosystem resilience, aligning with global trends in marine spatial planning that prioritize ecosystem service valuation (e.g., carbon sequestration and habitat protection) [13,15].

A key strength of this study is its emphasis on adaptive zoning mechanisms, such as buffer zones in Shanghai’s Chongming Island and storm surge-resilient wetland parks in Guangzhou’s Nansha New Area, which reflect a proactive response to climate-induced risks. These examples validate the necessity of incorporating dynamic thresholds (e.g., coastline development intensity indices) into zoning frameworks to strike a balance between ecological preservation and economic development. Furthermore, the integration of stakeholder participation mechanisms, exemplified by Hainan’s ecological compensation fund for coral reef restoration, aligns with international best practices in transboundary governance [18,28]. Such collaborative models not only reduce conflicts between industries and conservation but also promote long-term compliance with zoning regulations.

The choice of a 1 km grid for Shenzhen’s marine area aligns with China’s national marine monitoring standards (e.g., GB/T 17108-2006) and the resolution of Sentinel-2 satellite data (10–60 m), ensuring compatibility with existing environmental datasets such as water and sediment quality. Coarser resolutions, such as 5 km, would obscure critical coastal gradients, while finer resolutions (e.g., 100 m) would exceed the precision of socioeconomic datasets, such as port accessibility metrics. For Shenzhen’s 1145 km² marine area, a 1 km grid generates 1145 evaluation units, which is computationally manageable for entropy-weighted multi-criteria analysis. Using finer resolutions (e.g., 500 m) would quadruple computational demands without yielding proportional gains in planning granularity.

China’s territorial spatial planning guidelines recommend 1 km grids for provincial-scale zoning, ensuring consistency with higher-level planning frameworks. This allows vertical integration of local zoning into regional and national strategies. While the 1 km × 1 km grid adequately supported Shenzhen’s regional planning objectives, scale dependencies introduced measurable uncertainties in habitat-level assessments and coastal interface dynamics. Future work should focus on integrating adaptive grid systems and probabilistic uncertainty analysis to enhance the framework’s robustness across diverse coastal geomorphologies. This reflection strengthens the transparency of spatial decision making in marine zoning, aligning with global best practices in ecosystem-based management.

The proposed marine spatial zoning framework in Shenzhen aligns closely with China’s national “Territorial Spatial Planning (2020–2035)” through several key dimensions [70,71,72]. (1) It prioritizes ecological protection by emphasizing marine ecological spaces, with 38.53% of the coastline preserved in its natural state. This approach directly supports the national “ecological protection red line” strategy, aiming to protect 15% of territorial waters as ecological reserves by 2035. It also ensures that coastal cities maintain at least 35% natural shoreline retention, as stipulated by national policies. (2) The framework integrates multi-functional zoning, comprising ecological protection (26%), construction (34%), and biological resources (40%). This reflects the national “Three Zones and Three Lines” approach, where ecological control areas (22% of marine space) are designated as no-development zones, urban reserved areas offer flexible growth boundaries, and fisheries–tourism zones optimize agricultural production space. The framework uses an entropy-weighted index method to balance ecological (45% weight) and development (55%) factors, thereby meeting the national target of a 60:40 ecological–production–living space ratio. (3) The zoning incorporates climate resilience features such as storm surge buffer zones and dynamic shoreline development intensity indices. These adaptive measures support national requirements for coastal cities to establish 100–200 m climate adaptation buffers and ensure that artificial coastlines remain below 65% by 2035. With annual infrastructure growth constrained by carrying capacity limits (23.7%), the zoning strategy aligns with national climate resilience objectives. The zoning framework also emphasizes data-driven governance, employing 1 km² grid-based GIS analysis and machine learning tools to meet national mandates for marine environmental monitoring. These technologies enable real-time tracking of water quality parameters and ensure 90% monitoring coverage. The study’s digitization of Shenzhen’s 260.5 km coastline exceeds the national 1:10,000 scale mapping standard, facilitating precise and informed zoning decisions.

The integrated sea level rise and coastal zone adaptation zoning strategy for Shenzhen addresses the challenges of sea level rise and coastal management [73,74]. The western coast of Shenzhen, stretching from 22°32′ N to 22°40′ N, is 61.47% artificial, with reclamation and port construction obstructing tidal flows and sediment transport. This has led to a 42% reduction in intertidal habitats, based on GIS historical land use analysis, and has accelerated erosion rates. In contrast, the eastern coast remains 38.53% natural, with relatively stable sediment processes, although it faces future threats from sea level rise. Future sea level rise scenarios are based on localized IPCC RCP projections. Under the RCP 4.5 (moderate emissions) scenario, the sea level is expected to rise by 0.3–0.6 m by 2100, representing a medium risk. Under the RCP 8.5 (high emissions) scenario, the rise is projected to be 0.6–1.1 m by 2100, indicating a high risk [75]. Additionally, Shenzhen faces vertical land motion due to groundwater extraction, with a subsidence rate of 2–4 mm per year, which must be factored into sea level rise projections to adjust for absolute elevation loss [76].

Adaptation zoning thresholds will be adjusted accordingly in the future work. In ecological protection zones, new indicators include habitat elevation thresholds (such as mangrove elevation greater than 1.2 m) and coral bleaching risks. The ecological health status now includes predictions for habitat loss due to intertidal zone inundation. In Dapeng Bay, a dynamic buffer zone of 500–800 m has been piloted, prohibiting construction and allowing real-time monitoring of erosion and sedimentation data. For urban development zones, the shoreline development intensity is assessed using GIS-based inundation models to downgrade development scores for areas below the projected high tide lines. Additionally, a flood risk index integrates storm surge recurrence intervals, taking into account the increased frequency of 100-year return events. In biological resource utilization zones, the fisheries resource carrying capacity index simulates the impact of salinity intrusion on fish spawning grounds. Water quality standards are strengthened, particularly in estuarine areas, to monitor salinity gradients.

Dynamic zoning and model applications are also incorporated. Morphodynamic models will update marine protected area boundaries every 5–10 years to ensure sediment transport connectivity. Reclamation in low-lying areas, such as Qianhai and Shekou, will be restricted, and priority will be given to the redevelopment of existing land. The governance mechanisms include vertical integration with national and regional climate adaptation frameworks, such as the “National Climate Adaptation 2035” and the “Guangdong Coastal Zone Resilience Guidelines”. Horizontal coordination involves the establishment of a land development rights trading mechanism to compensate for economic losses in restricted development zones. This integrated approach aims to adapt Shenzhen’s coastal zone management to the challenges posed by sea level rise, ensuring sustainable development and ecological preservation.

The entropy-weighted multi-criteria framework reduces subjective bias by objectively quantifying ecological and developmental factors. However, it has limitations. The analysis relies on 2023–2024 water quality data (TOC, COD, nutrients) from fixed monitoring stations (e.g., Shenzhen Bay, West Bay) and a 1 km grid resolution. This static approach may overlook micro-scale coastal dynamics, such as sediment transport in western Shenzhen’s 61.47% artificial coastline, and temporal variations like algal bloom frequency, which require higher-resolution hydrodynamic modeling. Additionally, spatial validation gaps exist. While the framework identifies underutilized coastal areas, it lacks ground-truthing of suitability classifications against actual land-use patterns. For example, Nan’ao (114°28′ E) is categorized as high ecological suitability, but it faces tourism development pressures not fully captured by shoreline development intensity metrics. Similarly, Qianhai’s reclaimed zones (113°53′ E) show medium construction suitability despite existing port congestion issues. To enhance the reliability of the framework, several research areas are proposed for the future work: (1) multi-temporal validation, where suitability maps are compared against 5-year shoreline change rates derived from Sentinel-2 (10 m resolution); (2) stakeholder calibration by incorporating local fisheries catch data (e.g., Dapeng Bay) to verify biological resource zoning; (3) dynamic modeling by integrating storm surge inundation models to establish climate-resilient suitability thresholds.

Future research should also focus on improving the adaptability of zoning models through real-time data assimilation and scenario-based simulations. Strengthening interdisciplinary collaboration—such as combining hydrodynamic models with socioeconomic projections—could enhance the responsiveness of zoning plans to emerging threats like sea level rise. Furthermore, expanding participatory mechanisms to include marginalized coastal communities would ensure equitable resource allocation and promote stewardship. By addressing these gaps, marine spatial zoning can evolve into a dynamic tool for achieving the dual objectives of ecological integrity and sustainable development in coastal cities.

5. Conclusions

This study established a systematic approach to marine spatial planning that harmonized ecological integrity with urban development imperatives in coastal cities. The Shenzhen case demonstrated that suitability-driven zoning, grounded in multi-dimensional indicator analysis, effectively identified underutilized marine spaces for sustainable redevelopment while safeguarding critical ecosystems. The key findings revealed that: (1) spatially explicit suitability mapping enabled the differentiation of priority zones for conservation, resource utilization, and infrastructure expansion; (2) entropy-weighted evaluation methods mitigated subjectivity in zoning decisions by objectively quantifying ecological and socioeconomic trade-offs; and (3) integrated GIS modeling facilitated dynamic spatial optimization aligned with regional development goals.

The framework’s implementation in Shenzhen underscored three critical lessons for global coastal governance: First, the prioritization of ecological protection zones through quantitative habitat assessments prevented irreversible marine degradation. Second, adaptive zoning of construction spaces near existing infrastructure clusters minimized environmental footprints while maximizing economic returns. Third, marine biological resource zones required strict water quality thresholds and transportation accessibility criteria to ensure sustainable productivity.

By resolving land–sea planning fragmentation through science-based decision tools, this research advanced marine spatial planning theory and practice. Future applications should incorporate climate resilience metrics and participatory stakeholder mechanisms to enhance zoning adaptability. The study conclusively demonstrated that integrated spatial planning—supported by robust data systems and ecological prioritization principles—provided a viable pathway for coastal cities to achieve UN Sustainable Development Goals (SDGs) while maintaining economic competitiveness.