The Assessment of Land Suitability for Urban Expansion and Renewal for Coastal Urban Agglomerations: A Pilot Study of the Guangdong-Hong Kong-Macao Greater Bay Area

Abstract

Effectively and rationally allocating land resources, while coordinating urban expansion with internal renewal strategies, is crucial for achieving high-quality regional development in coastal urban agglomerations. Land-use suitability assessment (LSA) is a key method for coastal land-use planning, but it is primarily used to delineate ecological redlines or areas for urban expansion, often overlooking the spatial analysis needed for urban renewal. This is particularly critical in coastal urban agglomerations facing land scarcity and ecological fragility. Here, we combined land use and the Analytical Hierarchical Process (to consider stakeholder priorities) in a Minimum cumulative resistance model (MCRM) to determine suitable coastal urban growth and renewal based on a suite of 12 indicators relevant to development intensity and stock space. Application to the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) indicates a dominance of the Ecological Buffer Zone (70.5%), and the available stock space in the GBA comprises only 9.2% of the total area. Our modeling framework tailored different development strategies for different cities: Huizhou and Zhaoqing had space for urban expansion to varying degrees, while other cities were found to be suitable for urban renewal due to low stock space and high development intensity. Our modeling approach, incorporating stakeholder input and objective evaluation of geographic land-use information, can assist planners in improving ecological security while promoting high-quality developments in coastal areas.

1. Introduction

Over the last three decades, global urbanization has resulted in a significant increase in the demand for land in coastal areas [1,2]. The constant expansion of urban areas places immense pressure on land [3], leading to escalating conflicts across diverse land use types, particularly in coastal urban agglomerations. This has given rise to increasingly pressing eco-environmental concerns, including land resource shortages, ecosystem degradation, and landscape fragmentation [4,5,6]. Some of these challenges can be traced to the inadequate evaluation of land-use plans or policies implemented by governments [7]. In response, urban renewal strategies along with land-use suitability assessment (LSA) have been proposed as solutions to address these issues and assist in urban planning strategies [8,9].

The goal of land-use suitability assessment (LSA) is to identify the most appropriate spatial patterns for land-use layout optimization, offering policymakers crucial guidance on whether to prioritize urban expansion or urban renewal as part of their development strategies [10]. Much of the current focus in LSA research remains centered on ecological security and urban expansion [9,11], neglecting the spatial analysis necessary to inform urban renewal strategies. Meanwhile, studies on LSA in coastal urban agglomerations are relatively limited, with most focusing on individual cities or local scales, such as Ismailia Governorate in Egypt [12], Nong Khai in Thailand [13], and Beijing [14], and Nanchang [15] in China. Coastal urban regions present additional challenges due to their diverse land-use types, such as aquaculture land, mangroves, numerous seaports, and ecological protection areas, all of which impact spatial planning and ecological conservation efforts [16,17]. Thus, LSA in coastal urban agglomerations should incorporate considerations of the ocean’s economic contributions and the ecological impacts on landscape patterns.

In coastal urban agglomerations, where environmental sensitivity, economic pressures, and urban expansion need to be carefully balanced, the combination of AHP and MCRM offers a powerful approach to sustainable land use planning. Advances in spatial analysis tools have evolved LSA in coastal areas from simple overlays of physical factors to multi-dimension analysis [18]. Multi-Criteria Decision Analysis (MCDA), combined with stakeholder input on relative priorities, employs the Analytic Hierarchy Process (AHP) as a prevalent method for LSA [19,20]. Land use in coastal areas involves various complex factors, and AHP can organize these factors hierarchically, facilitating systematic analysis and comparison. At the same time, it can handle quantitative data and convert qualitative judgments into quantifiable indicators, making it suitable for the diverse and uncertain environment of coastal areas. However, AHP may potentially oversimplify the complexity of LSA by aggregating various inputs in a single/simple way, failing to effectively reflect the ecological processes in specific regions.

To address this limitation, the Minimum cumulative resistance model (MCRM) has been proposed as a complementary approach, incorporating spatial resistance to balance ecological protection with urban development expansion [21,22,23]. MCRM is an assessment method based on the simulation of landscape diffusion processes, emphasizing the cumulative effects of resource resistance over spatial distances, with its core focusing on optimizing the landscape pattern by considering ecological processes [24,25]. Although MCR can achieve spatial suitability assessment, its drawback lies in its inability to undertake multi-dimension analysis. Thus, combining AHP and MCRM allows for both the prioritization of key criteria (via AHP) and the consideration of spatial resistance to ecological protection (via MCR), ensuring a comprehensive land use assessment in coastal areas.

The Guangdong-Hong Kong-Macao Greater Bay Area (GBA) is one of the fastest-growing coastal urban agglomerations in China [26]. However, it faces significant challenges, including human-land conflict and ecological environment deterioration, which have limited high-quality development [16,27,28]. With many ecologically sensitive areas/reserves and uneven urban growth, historical development strategies that emphasized urban expansion are now hampered by the scarcity of land resources, compelling cities to shift towards urban renewal [29]. Bearing in mind these issues, we developed an LSA modeling framework specifically designed for coastal urban agglomerations, using the GBA as the study area. This framework encompasses a suitability index system that integrates seaports and coastal protection areas to emphasize the characteristics of coastal urban agglomerations. MCRM and AHP were used for LSA, while development intensity and available stock space were used to measure the development status in the GBA. This paper establishes a sound modeling basis for evaluating urban growth and renewal in the GBA, which can guide future high-quality development.

2. Study Area and Data

Located in southern China, the GBA is one of the world’s fastest-growing and most densely populated coastal urban agglomerations, containing 11 cities: Guangzhou (GZ), Shenzhen (SZ), Zhuhai (ZH), Zhaoqing (ZQ), Huizhou (HZ), Jiangmen (JM), Dongguan (DG), Foshan (FS), Hong Kong (HK), and Macao [16]. The GBA has experienced unprecedented urbanization and economic development, leading to significant land-use changes. With a population exceeding 90 million and a total GDP exceeding 11 trillion RMB in 2020, the GBA has become a vital engine for China’s economic growth.

However, rapid urbanization has intensified the demand for land resources, placing significant pressure on the coastal ecological environment. The GBA’s proximity to ecologically sensitive areas—such as coastal zones, wetlands, and forested mountains—has made the region particularly vulnerable to ecological degradation. With land resources becoming increasingly scarce and the ecosystem highly fragile, there is an urgent need for comprehensive ecological suitability assessments. Integrating ecological suitability assessments into regional planning is essential for maintaining the delicate balance between growth and conservation, ultimately ensuring that development in the GBA is both environmentally sustainable and economically resilient.

All the data are from the year 2020. Based on national and regional guidelines, expert knowledge, and available data, the selection of suitability factors included economic demands and natural factors. All the data were resampled to 30 m resolution. The location of the GBA is shown in Figure 1, and the data sources are described in

Table 1. Data sources for the GBA in 2020.

Name	Data Source
Ecological Control Area	Ministry of Natural Resources of the People’s Republic of China (http://g.mnr.gov.cn/, accessed on 1 May 2023)
Road Network	OpenStreetMap (https://www.openstreetmap.org/, accessed on 1 June 2024)
Boundary of GBA	Resource and Environmental Science Data Platform (https://www.resdc.cn/, accessed on 12 June 2023)
Land-use
NDVI
Terrain
DEM	Shuttle Radar Topography Mission (SRTM, https://earthexplorer.usgs.gov, accessed on 1 May 2023)
Population Density	WorldPop (https://www.worldpop.org, accessed on 1 May 2023)
GDP	Statistical yearbook of various cities in the GBA

.

3. Method

The LSA model for coastal urban agglomeration initially identifies pivotal factors influencing land suitability, including geographical conditions, the ecological environment, and human disturbances. Recognizing the economic and ecological conservation requirements of coastal urban agglomerations, the assessment criteria encompass seaports and coastal protection areas as indicators. Subsequently, the model develops a suitability assessment framework for coastal urban agglomerations, analyzing both available stock space and development intensity. The technical structure is illustrated in Figure 2.

The Minimum cumulative resistance model (MCRM) calculates the minimum process-related cost to overcome the resistance of all homogeneous or heterogeneous landscape units from a “source” or “sink”. It consists of three main parts: source, resistance plane, and resistance coefficient to another landscape:

$$MCR=fmin\sum _{j=n}^{i=m}{D}_{ij}\times R_i$$

(1)

where D_ij denotes the spatial distance from source j to landscape unit I; R_i denotes the resistance coefficient in the transition from landscape unit _i to source j.

Comparative resistance between ecological land and construction land was measured using the minimum cumulative resistance difference, as shown in Figure 3, where along the x-axis pathway, E denotes the ecological expansion source, U denotes the construction expansion source, and EC and UC denote the cumulative resistance expansion curves of ecological land and construction land, respectively. Between M and U, MCR for ecological expansion is greater; hence, construction land is more suitable for the M -> U pathway, and conversely, ecological land is favored between M -> E.

The MCR difference formula is as follows:

$${\mathrm{M}\mathrm{C}\mathrm{R}}_d={\mathrm{M}\mathrm{C}\mathrm{R}}_e-{\mathrm{M}\mathrm{C}\mathrm{R}}_u$$

(2)

where MCR_d represents the MCR difference between the two expansion processes, MCR_u and MCR_e for MCR of expansion of construction land and ecological land, respectively.

MCR_d < 0, indicates ecological land is easier to enlarge, and MCR_d > 0 indicates construction land is favored.

(1)

Source selections

Expansion sources are sources of ecological land that are rich in biodiversity and ecosystem services. In the GBA, such “ecological sources” include rivers, lakes, drinking water resources, significant wetlands, natural landscape reserves, and high mountains with rich natural resources. In principle, urban expansion must not encroach on this type of land in the GBA (Figure 4a). Also, construction land in 2020 is set as “sprawl sources” as shown in Figure 4b.

(2)

Resistance plane and coefficient

The resistance coefficient reflects the difficulty of each landscape unit in impeding the expansion of construction land or ecological land [31,32]. From the perspective of ecological processes, there is a competitive relationship between ecological land and construction land [33]. Generally, to ensure that ecological expansion and construction expansion can be compared on the same scale, a unified resistance evaluation system needs to be established. In this evaluation system, the grade assignment for the two is reversed.

In this study, the resistance plane is based on a thorough review of existing scientific research [22,34,35,36], with support from a panel of experts. All the resistance planes are divided into three main categories (primary indicators): Topography and geology, Ecological protection and Human disturbances. Each category is further subdivided into 12 subcategories, as shown in

Table 2. Valuation system for resistance factors for ecological and construction land.

		Ranking					Weight for Construction Land	Weight for Ecological Land
Resistance Plane of Construction Land		V	IV	III	II	I
Resistance Plane of Ecological Land		I	II	III	IV	V
Topography and geology	Slope	>25	10–25	5–10	2–5	<2	0.102	0.146
	Geological condition	Mountains, Depressions, Floodplains, Lakes	Hills, Terraces	/	/	Plain	0.051	0.073
	DEM	>160	120–160	80–120	40–80	1–40	0.073	0.105
Ecological protection	NDVI	>0.8	0.6–0.8	0.4–0.6	0.2–0.4	<0.2	0.053	0.074
	Land use	Water, Other forest, Aquaculture	Grassland	Economic forest	Cropland	Transportation, Industrial, Residential, Public	0.066	0.092
	Proximity to water (km)	<1	1–2	2–3	3–4	>4	0.094	0.131
	Ecological reserve	Ecological control land	-	-	-	Other area	0.053	0.074
	Coastal protection area	Guanghai Bay, Daya Bay	Coastal zone areas, Islands	Coastal zone forests	-	others	0.057	0.080
Human disturbances	Proximity to urban areas (km)	<0	0–0.5	0.5–1	1–1.5	>1.5	0.158	0.079
	Proximity to roads (km)	>4	3–4	2–3	1–2	0–1	0.123	0.062
	Population density	0–200	200–400	400–700	700–1000	>1000	0.093	0.046
	Port importance	0	1–2	2–5	5–8	>8	0.076	0.038

and Figure 5.

These include:

(1)

Slope: It affects risk management and resilience to coastal processes, potentially undermining the economic feasibility of the project.

(2)

Geological conditions: The geological composition of coastal areas influences how they will respond to sea-level rise. Areas with stable geological formations are more resistant to erosion and land loss, while low-lying areas composed of soft sediments are more vulnerable to flooding and permanent inundation.

(3)

DEM: It helps assess terrain, slope, flood risk, and elevation changes. Excessive elevation can hinder accessibility and increase flood risks in lower-lying areas.

(4)

NDVI: It measures vegetation health, helps identify green spaces, and supports ecological preservation. Higher NDVI values indicate denser, healthier vegetation, supporting ecological preservation.

(5)

Land use: Land use shapes existing development patterns, resource distribution, and environmental impact. It plays a key role in guiding sustainable planning by balancing ecological preservation and economic activities, especially in densely populated coastal regions.

(6)

Proximity to water: The presence of water is essential for the lifecycle in the coastal urban agglomeration.

(7)

Ecological reserve: It protects biodiversity, preserves critical habitats, and mitigates environmental degradation, which is crucial for future land planning.

(8)

Coastal protection area: It safeguards shorelines from erosion, mitigates storm impacts, preserves marine ecosystems, and ensures sustainable development by protecting vulnerable coastal environments from degradation and climate-related risks.

(9)

Proximity to urban areas: It influences access to infrastructure and services, promotes efficient land development, and reduces transportation costs.

(10)

Proximity to roads: It enhances accessibility, supports efficient transportation, and facilitates connectivity between urban areas.

(11)

Population density: It influences infrastructure demand, resource allocation, and environmental pressure in coastal areas.

(12)

Port importance: It facilitates international trade and promotes economic growth. Seaport importance is an index of the economic radiation intensity of ports across different cities.

In constructing resistance planes, suitability indicators were scored and ranked from I to V, where smaller values implied less resistance for the “source” to overcome in expansion. Since ecological land competes with construction land, indicators were assigned using the opposite-assignment method.

As previously mentioned, AHP is one of the most common and comprehensive methodologies, theorized by Saaty in 1980 [37]. Various papers about LSA employ the AHP to initiate the weights for their analyses [20,35,38]. It is suitable for solving problems where the factors can be organized in a hierarchical way [39]. The AHP calculation typically involves three main steps: first, constructing a hierarchical structure model; second, developing a judgment matrix; and finally, conducting a hierarchical ranking and consistency test [40]. The consistency test is a validation method used to determine whether the hierarchical ranking is feasible. At the same time, evidence of the validation can be found in the sensitivity analysis (Section 4.3), whose results clearly state that the factors have been addressed soundly.

In the AHP, experts were asked to define the importance of every sub-indicator on a scale from 1 (least important) to 9 (most important). The same was done for the three primary indicators (Topography and geology, Ecological protection, and Human disturbances). The study first set the weights for the resistance plane. Since the primary indicators differ in their level of importance for ecological land and construction land, different judgment matrices were used for calculating the weights of the primary indicators for ecological land and construction land, while the judgment matrix for the sub-indicators remained unchanged. The weights of each indicator, calculated using the AHP, are shown in Table 2, and the analysis results (

Table 3. Results of the consistency test.

Indicators for Weight Assignment	Consistency Ratio (CR)
Primary indicators of construction land	0.033
Primary indicators of ecological land	0.033
Sub-indicators of Topography and geology	0.033
Sub-indicators of Ecological protection	0.002
Sub-indicators of Human disturbances	0.004

) show that all CR values are less than 0.1, indicating they have passed the consistency test.

The method for delineating zoning thresholds often adopts the natural break method [31]. This study initially categorizes the region into two major classes based on the positive and negative relationships of the minimum cumulative difference: suitable for ecological land (difference less than 0) and suitable for construction land (difference greater than 0). Further subdivision of areas suitable for ecological and construction land is conducted using the natural break method. The zoning thresholds are outlined in

Table 4. Threshold ranges for land suitability zoning.

Landscape Type	Land Suitability Zoning	Threshold Range
Ecological land	ECZ: Ecological Control Zone	−67,927.9 to −23,161.5
	EBZ: Ecological Buffer Zone	−23,161.4 to 0
Construction land	SCZ: Suitable Construction Zone	0 to 6356.7
	PCZ: Prior Construction Zone	6356.8 to 23,395.4

, resulting in four categories: Ecological Control Zone (ECZ), Ecological Buffer Zone (EBZ), Suitable Construction Zone (SCZ), and Prior Construction Zone (PCZ).

4. Results

4.1. Spatial Characteristics of Suitable Areas for Urban Expansion and Ecological Protection

It can be observed in Figure 6 and Table 3 that the GBA primarily consists of EBZ, which accounts for 70% of the area, covering 39,289 km². This is followed by SCZ, which accounts for 22.8% or 12,706 km². The areas designated as ECZ and PCZ are comparatively smaller, measuring 2452 km² and 1282 km², respectively.

The EBZ is mainly located in the northwest and northeast parts of the GBA, with Zhaoqing, Huizhou, and Jiangmen each exceeding 7000 km² in area. Most EBZs are adjacent to SCZs, which act as buffer zones between suitable and unsuitable areas for urban development, such as water, cropland, and other forest. The cost of constructing infrastructure in such areas is relatively high. It is crucial to prioritize ecological preservation while ensuring rational utilization to prevent overdevelopment.

The SCZ is concentrated in peri-urban areas around original settlements and along main roads on either side of the Pearl River. These areas possess a favorable resource environment foundation and offer good conditions for economic and population agglomeration. They are the regions prioritized for urban expansion now and for a certain period in the future.

The PCZ is primarily concentrated in the core areas of various cities. These regions are typically characterized by dense populations and intense human activity, with a long history of urban development. The land use in these areas is deeply influenced by historical factors and does not exhibit a high degree of intensive utilization. Balancing high-quality development is challenging in these zones. Therefore, the current model of urban development urgently requires optimization and adjustment.

ECZ is sporadically distributed in the mountainous areas of northwest GBA and southern islands. These areas may either serve as important nature reserves with significant ecological importance, subject to strict policy controls, or are islands with scarce resources and high development challenges, unsuitable for large-scale development. Such regions should prioritize ecological protection unequivocally, enhance ecological construction, and harness ecological functions to promote the optimization of the regional ecological environment.

The proportion of suitability zoning in each city, as presented in Figure 7, indicates that over half of the GBA is predominantly categorized as EBZ. Zhaoqing, owing to its abundant forest and farmland, has the highest area of ECZ and EBZ. Guangzhou has the highest SCZ, while PCZ is highest in Shenzhen. The overlay of suitability zones for urban development with construction land, as shown in Figure 8, indicates significant urban expansion in the GBA, which nearly encompasses the entire area of the SCZ and the PZC. The areas suitable for urban expansion are mainly located in the northern part of Guangzhou, the southern part of Foshan, the bordering region between Huizhou and Shenzhen, as well as the surrounding areas of the built-up zone in Jiangmen.

Analyzing the proportion of each land use type in the suitability zones (Figure 9), we observe that water and other forests account for over 95% of the ECZ and EBZ. Only a portion of cropland, grassland, and other forests is suitable for urban expansion. The inclusion of some transportation land in the EBZ is due to the distribution of certain roads within the EBZ. In summary, the suitability results show significant differences in ecological environments and economic development among the cities within the GBA. The overall suitability zoning exhibits a distinct central-peripheral distribution pattern, decreasing from the core cities of Shenzhen, Dongguan, Guangzhou, and Foshan towards the northern and western inland areas of Guangdong.

4.2. LSA Heterogeneity Analysis

To better analyze development status, the total proportion of construction land in SCZ and PCZ, or “development intensity”, was used to characterize overall development, and available construction land for future expansion in SCZ and PCZ was characterized as “stock space”.

Nearly half of the GBA has a development intensity exceeding 70%, and the total stock space accounts for only 9.2% of the entire region (Figure 10 and Figure 11). Zhaoqing, Jiangmen, and Huizhou have large expandable potential in the future, but careful analyses are needed to adjust the regional development direction as the area is mainly mountainous. These areas are the three largest regions in the GBA with a notably uneven east-west development. Jiangmen and Huizhou have relatively low development intensity (<50%) and relatively high stock space (>1200 km²).

Macao, Shenzhen, Hong Kong, Zhongshan, and Dongguan need to seek regional development strategies that are intensive, compact, and efficient. The development intensity in these areas exceeds 80%, while their stock space is less than 400 km². These areas’ priority should be given to exploring already developed areas and promoting urban renewal to minimize the encroachment on the ecological space in these regions. Larger stock areas are often constrained by regional geography and ecological conditions, posing greater difficulties in development. Meanwhile, regions with higher levels of economic development exhibit significantly greater development intensity, suggesting limited available land for external expansion.

4.3. Sensitivity Analysis

The results of an MCDM analysis can be compromised by a large number of experts with divergent opinions on criteria selection and weight determination. This may stem from decision-makers’ incomplete awareness of their criteria preferences or uncertainty about the criteria’s nature and scale [41]. The validation of the results is, therefore, an important part of the analysis [42]. Sensitivity analysis is essential for the validation and calibration of numerical models, as it helps evaluate the robustness of the outcomes. In this study, sensitivity analysis was performed using a one-at-a-time approach, where the weights of each criterion were gradually increased while proportionally decreasing the weights of the others. This method allows for a clear assessment of the individual importance of each factor [43].

In this case, the weights of each criterion were increased by 2% at a time, up to a maximum of 20%. A total of 120 scenarios were analyzed. For each iteration, the changes in areas within each suitability zone are shown in

Table 5. Changes in the area belonging to an indicator (columns) in the “Slope” criterion after every 2% increase (rows). The suitability analysis row represents the values after the sensitivity analysis.

Weight Increment (%)	ECZ (km2)	EBZ (km2)	SCZ (km2)	PCZ (km2)
2%	2199	38,762	12,947	1481
4%	2184	38,769	12,970	1467
6%	2161	38,780	12,997	1453
8%	2208	38,721	12,989	1472
10%	2183	38,733	13,017	1457
12%	2158	38,746	13,044	1441
14%	2132	38,759	13,152	1346
16%	2105	38,773	13,102	1410
18%	2077	38,787	13,131	1394
20%	2092	38,761	13,173	1365

, reflecting the modifications made to the ‘Slope’ indicator. The row marked as “2%” signifies an incremental augmentation of 2% in the weight attributed to the indicator ‘Slope’. Notably, SCZ experiences an increase in count, while ECZ and PCZ show a decrease. It is essential to emphasize that the cumulative sum remains unchanged.

Due to the numerous indicators involved in the sensitivity analysis, only the more representative indicators are presented. The result shows that the ‘Geology condition’ is the least sensitive criterion, which produces minimal changes in SCZ. ‘Proximity to water’ is the most sensitive one, meaning that increases in its weight cause the most significant changes (Figure 12). Other sensitive criteria are ‘Proximity to urban areas’, ‘land use’, and ‘Proximity to roads’, all of which are high in the weight importance rank (Figure 13 and Figure 14). The least sensitive criteria, which produce minimal changes, are ‘NDVI’ and ‘Ecological reserve’. In conclusion, the prioritization of most criteria remained consistent, suggesting that weights were reliable even before the results were obtained. Therefore, the proposed model is deemed appropriate for the study.

5. Discussion

Coastal urban agglomerations face the dual challenges of land scarcity and ecological fragility [44]. Without proper regulation and control, coastal areas may sprawl disorderly, leading to ecological degradation and waste of land resources [7]. Urban renewal strategies, combined with land-use suitability assessments (LSAs), have been proposed as essential tools for land-use planning in coastal regions [45]. Previous studies have primarily focused on land-use suitability assessments (LSAs) applications for urban expansion and ecological security within localized or city-specific contexts [13,23,39], overlooking the spatial analysis required for urban renewal in these complex coastal regions. Thus, this study provides a comprehensive LSA framework that integrates the AHP and the MCRM to address the complex spatial planning needs of coastal urban agglomerations. Using the GBA as a case study, we demonstrate how this combined approach allows for a more holistic assessment of land-use suitability, balancing urban expansion and renewal with ecological preservation.

This study emphasizes the spatial challenges posed by diverse coastal land-use types based on 12 criteria, focusing on their impact on both coastal ecological protection and economic development [16,17]. Although Ref. [34] constructed an LSA framework for the GBA, it did not fully account for the economic contributions and ecological impacts of the unique landscapes and seaports within coastal zones. Our framework specifically addresses these factors, offering a more comprehensive assessment tool that incorporates both ecological and economic considerations in coastal urban agglomerations. In addition, the integration of the AHP and the MCRM demonstrates the value of combining qualitative decision-making tools with spatially explicit models to address the challenges posed by coastal environments [33,46]. This integrated approach is particularly important in coastal urban agglomerations like the GBA, where the competition between urban development and ecological protection is heightened due to the diverse land-use types and sensitive ecosystems [47].

Furthermore, by incorporating the spatial analysis of stock space and development intensity, our study shifts attention to the underexplored area of urban renewal, which is increasingly critical as land resources become scarcer in rapidly urbanizing regions like the GBA. The results reveal significant spatial heterogeneity in the availability of stock space across the GBA, with cities such as Zhaoqing, Jiangmen, and Huizhou showing greater potential for future expansion. In contrast, cities like Shenzhen, Hong Kong, and Macao exhibit high development intensity and limited stock space, highlighting the pressing need for urban renewal strategies. Our findings suggest that LSA can be a valuable tool for guiding renewal strategies. This aligns with recent calls in the literature for urban renewal in urban planning, which emphasize the importance of optimizing existing urban spaces to reduce environmental impacts and promote sustainable development [29,45].

Despite the contributions of this study, several limitations should be acknowledged. First, while the integration of AHP and MCRM provides a comprehensive framework for LSA, the reliance on AHP for multi-criteria decision-making may oversimplify the complexity of certain land-use interactions, as noted by previous studies [19,20]. Another limitation lies in the availability of data, particularly in accurately assessing the long-term ecological impacts of land-use changes in coastal regions. Future studies could benefit from dynamic datasets, better account for human-driven factors, and regularly update zoning classifications to align with evolving policy and environmental conditions. These limitations provide directions for future research, which could further refine the LSA framework to better account for the complex ecological, social, and economic interactions that influence land-use decisions in coastal urban areas.

6. Conclusions

In this study, we developed an LSA modeling framework specifically designed for coastal urban agglomerations, with the GBA as the study area. This framework encompasses a suitability index system consisting of 12 factors. AHP and MCRM were designed for LSA modeling, while development intensity and stock space were calculated for LSA heterogeneity analysis. This involves evaluating the development potential of regions and comparing suitability status outcomes for construction options among different cities in coastal urban agglomerations.

Results show that the GBA displays a distinct central-peripheral pattern. The GBA primarily consists of Ecological Buffer Zones (EBZs), accounting for 70.5%, mainly located in the northwest and northeast regions such as Zhaoqing, Huizhou, and Jiangmen. The focus on ecological preservation is paramount in these zones due to the high cost and challenges of developing infrastructure there. Surrounding these areas are Suitable Construction Zones (SCZs), covering 22.8% of the GBA, where urban expansion is prioritized due to favorable economic conditions and resource availability.

Prior Construction Zones (PCZs), concentrated in the core areas of the cities, face challenges due to dense populations and historical urbanization, making high-quality development difficult. There is a strong need for urban renewal in these zones. Ecological Control Zones (ECZs), found in mountainous regions and southern islands, prioritize ecological protection due to their critical environmental importance and developmental limitations. The overlay analysis indicates that areas for urban expansion primarily exist in the northern part of Guangzhou, southern Foshan, and areas between Huizhou and Shenzhen. The land suitability analysis reveals significant disparities between cities, with Shenzhen and Guangzhou’s core regions being more developed than other inland areas of the GBA.

Development intensity across the GBA shows that half of the region’s urban areas are over 70% developed, yet only 9.2% of the total land has potential for future expansion. Our modeling comparison of the GBA shows that Guangzhou is a central city for future development from north to east, which will strengthen regional services and radiation functions. Cities with higher economic development levels, such as Shenzhen, Dongguan, Macao, and Hong Kong, have limited space for expansion, highlighting the need for compact and efficient urban renewal strategies to minimize ecological disruption. Zhuhai and Zhongshan should focus on land resource utilization as overall stock space is limited (<300 km²). Foshan, adjacent to Guangzhou, has more than 700 km² of stock space, so it has the potential to expand. Cities like Zhaoqing, Jiangmen, and Huizhou have larger areas for potential expansion but face ecological and geographical constraints due to their mountainous terrain.

In conclusion, this study offers a significant contribution to the literature by advancing the methodology of LSA for coastal urban agglomerations and expanding its application to urban renewal. The combined use of AHP and MCRM offers a comprehensive framework for addressing the unique land-use challenges in these regions. Our modeling approach, incorporating stakeholders’ input and objective evaluation of complex geographic land-use information in coastal urban agglomerations of varying development stages, can assist planners to improve ecological security while promoting appropriate high-quality developments.