Optimizing Production–Living–Ecological Space Under Resource and Environmental Carrying Capacity Constraints: Evidence from Daye City, China

Abstract

Evaluating resource and environmental carrying capacity (RECC) serves as a fundamental approach for assessing regional environmental baselines and is widely applied in territorial spatial planning. Focusing on Daye City—a characteristic resource-exhausted city in Hubei Province—this study developed a comprehensive RECC evaluation system. By integrating the obstacle degree model, hotspot analysis, and Geodetector, we investigated the spatial differentiation mechanisms of RECC and the resulting production–living–ecological (PLE) spatial conflicts, ultimately proposing targeted optimization pathways. The core findings are as follows: (1) The RECC of Daye City exhibits pronounced spatial polarization and a distinct north–south gradient. (2) The spatial stress of industrial/mining land emerges as the primary obstacle (36.47%). Together with geological hazard risk and soil erosion sensitivity, it forms a core constraint chain. The highly significant hotspots of these factors strongly overlap in the north-central mining districts. (3) Geodetector analysis reveals robust bivariate and nonlinear enhancement effects among these core obstacle factors. This indicates that the cascading vicious cycle of mining disturbance, ecological degradation, and declining carrying capacity fundamentally underlies the constrained RECC in mining regions. (4) PLE spatial conflicts across the study area are dominated by production–ecological conflicts (47.73%), presenting a spatial pattern that heavily couples with the polarized obstacle zones. Based on these findings, this study proposes differentiated regulation strategies centered on mitigating mining-induced stress and interrupting the cascading transmission of disaster risks. These strategies aim to restructure and optimize the territorial spatial pattern, providing robust quantitative decision support for the sustainable transformation of similar resource-exhausted cities.

1. Introduction

Against the backdrop of the national “dual carbon” goals and the comprehensive promotion of high-quality development, optimizing territorial spatial structure has become a key pathway for achieving regional green and low-carbon transitions. The 69 national-level resource-exhausted cities officially designated by the State Council remain highly carbon-intensive due to prolonged extensive resource extraction, resulting in severe conflicts among human activities, land use, and mining development [1,2]. Resource and environmental carrying capacity (RECC), as a core indicator reflecting regional resource availability and the carrying limits of human activities, serves as a critical basis for resolving such conflicts and forging a coordinated development–conservation pattern [3,4]. China has issued a series of policy initiatives, including the National Plan for Sustainable Development of Resource-Based Cities (2013–2020) and the 14th Five-Year Implementation Plan for Promoting High-Quality Development in Resource-Based Regions, which explicitly designate the cultivation of alternative industries, the advancement of mine ecological restoration, and the optimization of territorial spatial patterns as core tasks. Empirical evidence from cities that have achieved substantial transformation—such as Tongling in Anhui Province and Jingdezhen in Jiangxi Province—shows that scientific diagnosis and systematic restructuring of production–living–ecological (PLE) spaces offer an effective pathway to sustainable transformation [5]. Designated in 2008 as one of China’s first national resource-exhausted cities, Daye has endured severe ecological degradation from decades of intensive copper mining; at the same time, its traditional heavy and chemical industries face mounting emission-reduction pressures. Its transformation trajectory thus stands as a highly representative case in this research domain. In this context, the present study selects Daye City as the case area to examine optimization pathways for PLE spaces under RECC constraints. The aim is to alleviate bottlenecks in its territorial spatial development while offering both methodological insights and actionable insights for the green transformation of comparable resource-exhausted cities.

Research on RECC has expanded beyond ecology into multiple disciplines, accompanied by increasingly refined evaluation frameworks [6,7,8,9,10]. Existing studies have incorporated multiple approaches into RECC assessments, such as system dynamics, ecological footprint analysis, and big data-driven early warning techniques, and the mutation progression method [11,12,13,14]. For example, Khan developed a system dynamics-based framework to evaluate water and soil resource carrying capacity at the watershed scale [15]; similarly, Pravitasari quantified the carrying status of metropolitan areas in Java Island using an ecological footprint model [16]. Tong et al. further developed a big data-based dynamic monitoring system, while Peng et al. proposed a “pressure–support–regulation” framework [17,18]. With the establishment of China’s territorial spatial planning system, the concept of PLE spaces has gradually evolved into three functional categories—ecological, agricultural, and urban spaces—which underpin zoning control and land-use regulation [19]. This framework is increasingly adopted to address spatial conflicts and to optimize territorial spatial patterns [20,21,22]. Existing studies have applied the PLE perspective in various contexts. For example, Bole et al. developed a resource–ecological carrying capacity evaluation system for western Jilin Province [23], Jiang et al. explored how land-use optimization can balance urban expansion and resource-environmental constraints [24,25]. Wang assessed the evolutionary dynamics of production–living–ecological space carrying capacity in tourism-centric traditional villages, demonstrating the framework’s applicability to culturally sensitive rural areas [26]. Yet, existing research on PLE space reconfiguration under RECC constraints has mostly focused on macro scales (e.g., provinces or urban agglomerations), with limited studies at the county level—especially for resource-exhausted cities. Taking Daye City as a case study, this study reconstructs PLE spaces based on RECC evaluation and seeks to identify spatial optimization pathways for the sustainable transformation of resource-exhausted cities.

2. Materials and Methods

2.1. Study Area

Daye City (114°31′ E–115°20′ E, 29°40′ N–30°15′ N) is located in southeastern Hubei Province and is an important component of the Wuhan Metropolitan Area (Figure 1). It covers a land area of 1556.98 km² and administers five subdistricts and eleven towns and townships. The south is dominated by the residual hills of the Mufu Mountains, forming a fundamental ecological barrier. The central and northern areas are characterized by flat land and dense river–lake systems including Bao’an Lake and Daye Lake, with urban built-up areas and cultivated land mainly concentrated here. Influenced by a subtropical monsoon climate, the study area has adequate hydrothermal conditions suitable for local agriculture and ecological stability. In 2024, Daye’s GDP reached 93.1 billion CNY, with a solid socioeconomic base, and occupies a strategic position in the Wuhan–Ezhou–Huangshi–Huanggang Urban Cluster. As a typical resource-exhausted city, Daye is undergoing a process of socio-ecological transformation while facing the intertwined challenges of resource depletion, ecological degradation, and industrial restructuring.

2.2. Data Sources and Preprocessing

The datasets employed in this study comprise physical geographic data, socio-economic statistics, and fundamental geospatial information. All raw data underwent a systematic preprocessing workflow, including data cleaning, multi-source spatial alignment, and thematic integration, as detailed below:

(1)

Conventional statistical and remote sensing data: Statistical and remote sensing data were derived from authoritative national scientific data platforms and government statistical yearbooks (Table 1). Following missing value filling, outlier removal and coordinate consistency processing, multi-source datasets were spatially matched using spatial interpolation.

(2)

Geospatial big data: A location accessibility index was derived from commuting time estimates based on road network data acquired through the Baidu Maps API, with a 3 km × 3 km grid and weighted interpolation. The social security level was evaluated by the spatial accessibility of public service facilities. POI data for medical, financial and other public service facilities were collected using Python 3.13, and accessibility was calculated via the Dijkstra shortest-path algorithm.

(3)

Thematic data: Relevant data were collected, including the “Three Zones and Three Lines”, mining rights distribution, river and lake protection boundaries, drinking water source protection zones, flood control zones, forestry development plans, land supply plans, and industrial park development plans, to improve the regional adaptability of the evaluation.

Table 1. Description of research data and sources.

Data Category	Data Name	Data Source	Accuracy/Resolution
Ecological Environment	Meteorological Data (Precipitation/Accumulated Temperature, 2015)	Resource and Environment Science and Data Center, Chinese Academy of Sciences (RESDC, CAS)	1 km
	Land Cover Data (GlobeLand30, 2020)	China Knowledge Centre for Engineering Sciences and Technology (CKCEST)	30 m
	Remote Sensing Imagery (Landsat 8 OLI, 2018)	Geospatial Data Cloud	30 m
	PM2.5 Concentration (2018)	Data Sharing Service System of the Big Earth Data Science Engineering Program (CASEarth)	1 km
	Potential Geological Hazard Sites (2020)	Daye Municipal Bureau of Natural Resources and Planning	Vector Point Data
Land Resources	Digital Elevation Model (DEM, 2020)	Geospatial Data Cloud (GDEMV2)	30 m
	Soil Properties (Silt Content/Organic Matter, 2009)	National Tibetan Plateau Data Center (Harmonized World Soil Database, HWSD)	1 km
	Surface Effective Radiation (Multi-Year Average, 1971–2000)	National Ecological Science and Technology Resource Service Platform	/
Social Development	Basic Geospatial Information (Administrative Boundaries/Road Networks)	Daye Municipal People’s Government, OpenStreetMap (OSM)	Vector
	Socio-Economic Statistics (GDP/Population, 2021)	Encyclopedia of Administrative Divisions of the People’s Republic of China, Seventh National Population Census of the People’s Republic of China	Township level
	Social Sensing Data (Commuting/POI Data)	Baidu Maps Open API, Web Crawling and Secondary Calculation	/

Note: Italic type denotes the full title of an official published reference book.

Table 1. Description of research data and sources.

Data Category	Data Name	Data Source	Accuracy/Resolution
Ecological Environment	Meteorological Data (Precipitation/Accumulated Temperature, 2015)	Resource and Environment Science and Data Center, Chinese Academy of Sciences (RESDC, CAS)	1 km
	Land Cover Data (GlobeLand30, 2020)	China Knowledge Centre for Engineering Sciences and Technology (CKCEST)	30 m
	Remote Sensing Imagery (Landsat 8 OLI, 2018)	Geospatial Data Cloud	30 m
	PM2.5 Concentration (2018)	Data Sharing Service System of the Big Earth Data Science Engineering Program (CASEarth)	1 km
	Potential Geological Hazard Sites (2020)	Daye Municipal Bureau of Natural Resources and Planning	Vector Point Data
Land Resources	Digital Elevation Model (DEM, 2020)	Geospatial Data Cloud (GDEMV2)	30 m
	Soil Properties (Silt Content/Organic Matter, 2009)	National Tibetan Plateau Data Center (Harmonized World Soil Database, HWSD)	1 km
	Surface Effective Radiation (Multi-Year Average, 1971–2000)	National Ecological Science and Technology Resource Service Platform	/
Social Development	Basic Geospatial Information (Administrative Boundaries/Road Networks)	Daye Municipal People’s Government, OpenStreetMap (OSM)	Vector
	Socio-Economic Statistics (GDP/Population, 2021)	Encyclopedia of Administrative Divisions of the People’s Republic of China, Seventh National Population Census of the People’s Republic of China	Township level
	Social Sensing Data (Commuting/POI Data)	Baidu Maps Open API, Web Crawling and Secondary Calculation	/

Note: Italic type denotes the full title of an official published reference book.

2.3. Evaluation Model of RECC

2.3.1. Index System Construction and Weighting Method

Following the principles of systematicity, scientific rigor, and operability, we constructed a RECC evaluation index system. This framework integrates territorial function theory with the unique characteristics of Daye City as a resource-exhausted city, drawing upon the official Chinese Guidelines for the Evaluation of Resource and Environmental Carrying Capacity and Suitability of Territorial Space Development (Trial), as well as the evaluation frameworks established by Guo, Ozsahin et al. [27,28,29]. The system comprises three subsystems (ecological environment, agricultural production, and socio-economic development) and 27 specific indicators (detailed in Table 2). To provide uniform spatial analytical units for RECC evaluation and spatial statistical analyses, a 1 km × 1 km fishnet grid was generated across Daye City, and the centroid of each grid cell was extracted as a sampling point. After excluding cells located outside the study boundary or lacking complete attribute information, 1331 valid points (from an initial 1557 points) were retained for further analysis. To determine the relative importance of each indicator, a combined weighting approach using the Analytic Hierarchy Process (AHP) and the entropy weight method was adopted. This hybrid method couples the subjective judgment of experts with the objective information embedded in spatial data, overcoming the inherent limitations of single weighting techniques [30,31,32].

(1)

Indicator Standardization: To eliminate differences in dimensions and magnitudes among indicators, all variables were standardized using the min–max normalization method. Positive and negative indicators were processed according to Equations (1) and (2), respectively.

$$X={\displaystyle \frac{x-Min\left(x\right)}{Max\left(x\right)-Min\left(x\right)}}$$

(1)

$$X={\displaystyle \frac{Max\left(x\right)-x}{Max\left(x\right)-Min\left(x\right)}}$$

(2)

(2)

Subjective Weight Determination: To ensure the reliability and regional applicability of the evaluation, a panel of 37 experts and local practitioners from fields such as territorial spatial planning, ecological assessment, and agricultural resources was selected. The expert questionnaire utilized scoring guidelines based on a series of national standards, including the Code for Classification of Urban Land Use and Planning Standards of Development Land (GB 50137-2011), Regulations for Gradation on Agricultural Land Quality (GB/T 28407-2012), and Ecosystem Assessment (GB/T 43678-2024) [33,34,35]. After excluding samples with logical inconsistencies, 31 valid questionnaires were used for weight calculation.

The Saaty 1–9 scale was employed to construct the pairwise comparison matrix $A={\left[a_{ij}\right]}_{n\times n}$, where $a_{ij}=1∕a_{ji}$ and $a_{ii}=1$. The geometric mean method was used to aggregate the independent scores from the 31 experts into a comprehensive group decision matrix:

$$a_{ij}={\left({\displaystyle \prod _{k=1}^N}{a}_{ij}^{\left(k\right)}\right)}^{1/N}$$

(3)

Subsequently, the square root method was applied to calculate the local weight $w_i$ of each indicator, followed by normalization:

$$w_i={\displaystyle \frac{{\left({\Pi }_{j=1}^n{a}_{ij}\right)}^{1/n}}{{\Sigma }_{i=1}^n{\left({\Pi }_{j=1}^n{a}_{ij}\right)}^{1/n}}}$$

(4)

To verify the logical consistency of expert judgments, consistency tests were conducted on the aggregated judgment matrices. By calculating the maximum eigenvalue (${\lambda }_{max}$), the Consistency Index (CI) and Consistency Ratio (CR) were obtained using $CI=\left({\lambda }_{max}-n\right)∕\left(n-1\right)$ and $CR=CI∕RI$, where RI denotes the Random Index. A judgment matrix was considered acceptable when $CR<0.1$. All aggregated matrices satisfied this criterion, indicating satisfactory consistency of expert judgments. The corresponding pairwise comparison matrices and consistency statistics are provided in Appendix A (

Table A1. Ecological environment subsystem AHP judgment matrix and consistency test results.

	Ecological Services	Ecosystem Risks	Ecosystem Constraints
Ecological services	1.000	0.512	0.327
Ecosystem risks	1.953	1.000	0.491
Ecosystem constraints	3.058	2.037	1.000
λmax = 3.0098	CI = 0.0049	CR = 0.0084	RI = 0.58

,

Table A2. Agricultural production subsystem AHP judgment matrix and consistency test results.

	Agro-Climatic Suitability	Agricultural Water Suitability	Agricultural Land Suitability
Agro-climatic suitability	1.000	0.487	0.213
Agricultural water suitability	2.053	1.000	0.342
Agricultural land suitability	4.695	2.924	1.000
λmax = 3.0042	CI = 0.0021	CR = 0.0036	RI = 0.58

and

Table A3. Socio-economic development subsystem AHP judgment matrix and consistency test results.

	Urban Construction Suitability	Social Development Level	Social Security Level
Urban construction suitability	1.000	0.524	0.497
Social development level	1.908	1.000	1.315
Social security level	2.012	0.760	1.000
λmax = 3.0083	CI = 0.0042	CR = 0.0072	RI = 0.58

). In addition, Kendall’s coefficient of concordance (W) was calculated to assess inter-rater reliability among the 31 experts. The results indicated significant agreement among the experts (W = 0.732, p < 0.001), further supporting the robustness of the AHP weighting process.

(3)

Objective Weight Determination: To mitigate the potential decision bias inherent in purely subjective weighting, the entropy weight method was introduced to determine the objective weights based on the dispersion of the indicator data.

First, the proportion $P_{ij}$ of the $i$-th evaluation unit under the $j$-th indicator was calculated:

$$P_{ij}={\displaystyle \frac{Y_{ij}}{{\Sigma }_{i=1}^m{Y}_{ij}}}$$

(5)

where $Y_{ij}$ denotes the data processed by min–max normalization, and $m$ is the total number of evaluation units (i.e., spatial grids in Daye City).

Next, the information entropy $E_j$ of the $j$-th indicator was computed:

$$E_j=-{\displaystyle \frac{1}{\mathit{ln}m}}{\displaystyle \sum _{i=1}^m}{P}_{ij}\mathit{ln}{P}_{ij}$$

(6)

(Note: When $P_{ij}=0$, $P_{ij}\ln P_{ij}=0$ is defined as 0 to prevent invalid calculations.)

Finally, the objective weight $W_j^{″}$ of the $j$-th indicator was obtained:

$$W_j^{″}={\displaystyle \frac{1-E_j}{{\Sigma }_{j=1}^n\left(1-E_j\right)}}$$

(7)

where n represents the total number of evaluation indicators (27 in total), and $1-E_j$ indicates the information utility value of the $j$-th indicator.

(4)

Combined Weight Calculation: A normalized multiplicative synthesis method was employed to calculate the final comprehensive weights. This approach amplifies core indicators that possess significant weights both subjectively and objectively, thereby enhancing the discriminatory power of the evaluation results [36,37]. The calculation is as follows:

$$W_j={\displaystyle \frac{W_j^{\prime }\times W_j^{″}}{{\Sigma }_{j=1}^n\left(W_j^{\prime }\times W_{\dot{J}}^{″}\right)}}$$

(8)

where $W_j^{\prime }$ is the subjective weight (AHP), $W_j^{″}$ is the objective weight (entropy method), and $W_j$ represents the final comprehensive weight of the j-th indicator.

Table 2. RECC evaluation index system.

Subsystem	Criterion Layer	Indicator Layer	Comprehensive Weight
Ecological Environment	Ecological services	Water conservation	0.2550
		Soil and water conservation	0.5218
		Windbreak and sand fixation	0.2232
	Ecosystem risks	Soil erosion sensitivity	0.4753
		Rocky desertification sensitivity	0.5247
	Ecosystem constraints	Air pollution	0.1538
		Geological hazard risk	0.8462
Agricultural Production	Agro-climatic suitability	Surface effective radiation	0.3274
		Photo-thermal productivity	0.4011
		Arable land aspect	0.2715
	Agricultural water suitability	Regional drought index	0.4006
	Agricultural land suitability	Arable land elevation	0.1780
		Arable land slope	0.3435
		Soil organic matter content	0.2833
		Soil silt content	0.1952
Socio-Economic	Urban construction suitability	Land construction suitability	0.1103
		Location advantage index	0.1836 0.7061
		Spatial stress of industrial/mining land	0.1189
	Social development level	Transport accessibility	0.2661
		Population density	0.2725
		Regional gross industrial output	0.2886
		Regional consumption level	0.054
		Regional land-use intensity	0.2265
	Social security level	Medical service accessibility	0.2462
		Industrial distribution density	0.3145
		Service facility level	0.2128
		Cultural and recreational facility level	0.2550

Table 2. RECC evaluation index system.

Subsystem	Criterion Layer	Indicator Layer	Comprehensive Weight
Ecological Environment	Ecological services	Water conservation	0.2550
		Soil and water conservation	0.5218
		Windbreak and sand fixation	0.2232
	Ecosystem risks	Soil erosion sensitivity	0.4753
		Rocky desertification sensitivity	0.5247
	Ecosystem constraints	Air pollution	0.1538
		Geological hazard risk	0.8462
Agricultural Production	Agro-climatic suitability	Surface effective radiation	0.3274
		Photo-thermal productivity	0.4011
		Arable land aspect	0.2715
	Agricultural water suitability	Regional drought index	0.4006
	Agricultural land suitability	Arable land elevation	0.1780
		Arable land slope	0.3435
		Soil organic matter content	0.2833
		Soil silt content	0.1952
Socio-Economic	Urban construction suitability	Land construction suitability	0.1103
		Location advantage index	0.1836 0.7061
		Spatial stress of industrial/mining land	0.1189
	Social development level	Transport accessibility	0.2661
		Population density	0.2725
		Regional gross industrial output	0.2886
		Regional consumption level	0.054
		Regional land-use intensity	0.2265
	Social security level	Medical service accessibility	0.2462
		Industrial distribution density	0.3145
		Service facility level	0.2128
		Cultural and recreational facility level	0.2550

2.3.2. Obstacle Degree Model

To accurately identify the core obstacle factors restricting the improvement of RECC in Daye City, the obstacle degree model was introduced for quantitative diagnosis. This model quantifies the inhibitory effect of each indicator on RECC enhancement through three core parameters: factor contribution degree, index deviation degree, and obstacle degree [38,39]. The specific calculation formulas are specified as follows:

(1)

Factor Contribution Degree $F_j$: This indicator represents the contribution of a single indicator to the overall evaluation objective, with its value equal to the comprehensive weight of the corresponding single indicator:

$$F_j=W_j$$

(9)

where $W_j$ is the comprehensive weight of the j-th indicator, which is fully consistent with the calculation results of the combined weighting method described above.

(2)

Index Deviation Degree $I_{ij}$: This indicator reflects the gap between the actual value of a single indicator and the optimal value for carrying capacity, with its value defined as the difference between 1 and the normalized value of the indicator:

$$I_{ij}=1-X_{ij}$$

(10)

where $X_{ij}$ is the normalized value of the j-th indicator for the i-th evaluation unit, which is consistent with the results of the min–max normalization process detailed in the previous section.

(3)

Obstacle Degree $O_{ij}$: This indicator quantifies the inhibitory effect of a single indicator on the improvement of carrying capacity. Meanwhile, the comprehensive obstacle degree of categorical indicators can be calculated via weighted averaging to identify the core restricting factors:

$$O_{ij}={\displaystyle \frac{I_{ij}\times F_j}{{\Sigma }_{j=1}^n{I}_{ij}\times F_j}}\times 100\%$$

(11)

where $O_{ij}$ is the obstacle degree of the j-th indicator for the i-th evaluation unit; a larger value indicates a stronger inhibitory effect on the improvement of carrying capacity; the summation ranges from j = 1 to n = 27 (the total number of evaluation indicators in this study).

2.3.3. Calculation of Key Parameters

(1)

Soil and Water Conservation Function: The soil and water conservation capacity was evaluated based on slope gradient, slope length, fractional vegetation cover, and the percentages of soil clay, silt, sand, and organic carbon [40], as expressed in Equation (10):

$$\mathrm{ES}\mathrm{s}\mathrm{w}\mathrm{c}= \mathrm{K}\times \mathrm{L}\times \mathrm{S}\times \left(1-\mathrm{C}\right)$$

(12)

where $\mathrm{ES}swc$ represents the evaluated soil and water conservation capacity; K is the soil erodibility factor; L and S are the topographic factors for slope length and gradient, respectively; and C is the vegetation cover factor.

The soil erodibility factor (K) characterizes soil sensitivity to water erosion, which significantly correlates with soil organic matter content and particle composition [41,42]. It is calculated using Equation (11):

$$\begin{gathered} \mathrm{K}=\left[-0.01383+0.51575\mathrm{Kepic}\right]\times 0.1317 \\ K\mathrm{epic}=\{0.2+0.3\exp [-0.0256\mathrm{m}s(1 - \mathrm{msilt}/100)\left]\right\}\times [m\mathrm{silt}/(mc+m\mathrm{silt}){]}^{0.3}\times \{1 - 0.25\mathrm{orgC}/[\mathrm{orgC} \\ +\exp (3.72 - 2.95\mathrm{orgC}\left)\right]\}\times \{1 - 0.7(1 - \mathrm{m}s/100)/\left\{\right(1 - \mathrm{m}s/100)+\exp [-5.51+22.9(1 - \mathrm{m}s/100)\left]\right\}\} \end{gathered}$$

(13)

where $mc$, $m\mathrm{silt}$, and $\mathrm{m}s$ denote the percentages of soil clay, silt, and sand, respectively; and $\mathrm{orgC}$ is the percentage of soil organic carbon.

The vegetation cover factor (C) reflects the mitigating effect of vegetation on soil erosion. Following the N-SPECT model parameterization, the C value was set to 0 for paddy fields and wetlands, 0.01 for construction land, and 0.7 for barren land. For rainfed cropland, the C value was calculated using Equation (12):

$$C_{\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{d}} = 0.221 - 0.595\mathrm{logc}i$$

(14)

where $C_{\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{d}}$ is the vegetation cover factor for rainfed cropland, and $\mathrm{c}i$ indicates the fractional vegetation cover.

(2)

Medical Security Level: Travel times from each sampling point to nearby hospitals and clinics were retrieved via the Baidu Maps Open API using Python scripts [43,44]. The minimum travel time was identified and transformed based on Equations (13) and (14). The final spatial distribution was then generated through spatial interpolation and normalization processing [45,46,47].

$$MSi=\left\{\begin{array}{ccc}1-{\displaystyle \frac{Ti}{3000}}& ,& 0 < Ti \le 3000\\ 0& ,& Ti > 3000\end{array}\right.$$

(15)

$$Ti=M\mathrm{in}\left(ti1, ti2, ti3, ti4, \dots ,tij\right)$$

(16)

where $MSi$ is the transformed value of the minimum travel time for sampling point i; $Ti$ represents the absolute shortest travel time (in seconds) from sampling point $i$ to the nearest hospital; and $tij$ is the travel time (in seconds) from sampling point $i$ to hospital $j$.

2.3.4. RECC-Driven Spatial Optimization Method

The carrying capacities of the ecological environment, agricultural production, and socio-economic subsystems align with the functional intensities of ecological, agricultural, and construction spaces, respectively. These capacities are highly consistent with the functional attributes of PLE spaces and constitute the fundamental baseline conditions for spatial optimization. To characterize the mapping relationship between carrying capacity subsystems and PLE spatial functions, a three-dimensional identification model of RECC was developed (Figure 2).

The evaluation values of the three subsystems are used as coordinate axes. Following the principle of prioritizing ecological protection while ensuring production, the model enables multi-dimensional spatial partitioning through suitability classification of each subsystem, thereby delineating PLE spaces with different functional orientations. The main steps are as follows: (1) Normalize the evaluation values of each subsystem; (2) classify the values according to carrying capacity levels using the Jenks natural breaks method [48,49] (

Table 3. Suitability classification for RECC.

Carrying Capacity Type	Functional Suitability Classification	Value Range
Ecological Environment (EECC)	Important ecological function zone	EECC ≤ 0.52
	Moderately important ecological function zone	0.52 < EECC ≤ 0.70
	Non-important ecological function zone	0.70 < EECC
Agricultural Production (ACC)	Highly suitable area for agricultural production	ACC ≤ 0.47
	Moderately suitable area for agricultural production	47 < ACC ≤ 0.71
	Unsuitable area for agricultural production	0.71 < ACC
Socio-Economic (SCC)	Highly suitable area for socio-economic development	SCC ≤ 0.41
	Moderately suitable area for socio-economic development	0.41 < SCC ≤ 0.62
	Unsuitable area for socio-economic development	0.62 < SCC

); (3) delineate ecological, agricultural, and living functional spaces based on the classification results; (4) compare the results with existing PLE spaces derived from current land use to identify spatial conflicts and optimize the spatial pattern.

3. Results

3.1. Evaluation Results of RECC

3.1.1. Evaluation of the Ecological Environment Subsystem

The ecological environment subsystem includes three dimensions: ecosystem services, ecosystem risks, and ecosystem constraints. High-value ecosystem service areas are concentrated in the foothills of the Mufu Mountains in southern Daye City. Covered mainly by forest land, this region provides critical ecological functions and serves as the primary ecological barrier for the city (Figure 3a). Elevated ecosystem risks occur mainly in the cultivated transition zone between the southern mountains and the plains. These areas are highly susceptible to soil erosion and rocky desertification (Figure 3b). As for ecosystem constraints, Daye City—a traditional mining-dependent resource city—experiences elevated PM_2.5 concentrations in its northern urban area owing to industrial agglomeration and traffic emissions. Mining zones in Jinhu Subdistrict, Jinshandian Town, and Chengui Town are also highly prone to geological hazards, which exert particularly strong constraints on local ecological carrying capacity (Figure 3c). Integrating the three components produces the comprehensive evaluation map of the ecological environment subsystem (Figure 3d). Based on 1331 valid sampling points, the coefficient of variation in the subsystem was 0.234. Overall, the ecological carrying capacity of Daye City displays pronounced spatial heterogeneity: functional zones exhibit patch- and band-like distributions, with marked gradients along both north–south and east–west axes.

3.1.2. Evaluation of the Agricultural Production Subsystem

The agricultural production subsystem consists of three dimensions: agro-climatic suitability, water resource suitability, and agricultural land suitability. High agro-climatic suitability values are widely distributed across northern Daye City. These areas possess higher accumulated temperatures above 0 °C than the southern mountains and benefit from abundant solar and thermal resources, creating optimal climatic conditions for farming (Figure 4a). Water resource suitability closely tracks the distribution of water networks in the northern and southwestern parts of the city. High regional moisture indices, abundant precipitation, and low evaporation rates—coupled with a dense network of rivers and lakes—provide reliable water sources for agricultural irrigation (Figure 4b). Agricultural land suitability is mainly constrained by topography and soil properties. Highly suitable zones occupy the plains and gentle hills of the north and southwest, where slopes are generally below 15°. Although soil silt and organic matter contents vary locally, these areas exhibit strong overall suitability for agriculture. In contrast, low-suitability zones are mainly located in the southern mountainous areas, where slopes exceed 15°, soils exhibit relatively low fertility, and the terrain and soil conditions do not support large-scale agricultural production (Figure 4c). With a coefficient of variation of 0.477—the highest among the three subsystems—this subsystem displays the most pronounced spatial heterogeneity in the study area. Integrating the three dimensions shows that the core agricultural carrying zone lies in the area extending from the northern to the southwestern part of the city, forming the dominant agricultural space in the city (Figure 4d).

3.1.3. Evaluation of the Socio-Economic Subsystem

The socio-economic subsystem comprises three dimensions: urban construction suitability, social development level, and social security level. Urban construction suitability exhibits a pronounced north–south gradient. Highly suitable zones are concentrated in Dongyue Subdistrict and its surrounding areas, with suitability gradually decreasing outward along the Daqing–Guangzhou, Wuhan–Yangxin, and Qichun–Jiayu expressways (Figure 5a). The level of social development is strongly driven by industrial and population agglomeration, peaking in the northern core subdistricts while lagging in the southern townships (Figure 5b). Overall, the level of social security and public services in the study area is generally favorable. The high-value areas of industrial and service facilities form a dual-core spatial structure, with Dongyue Subdistrict Office as the primary core and Huandiqiao Town as the secondary growth pole, which radiates outward along major transportation arteries (Figure 5c). With a coefficient of variation of 0.413, this subsystem is characterized by significant core-periphery differentiation. Integrating these dimensions, the socio-economic carrying capacity of Daye City demonstrates a clear decreasing trend from the central urban area toward the southern and western peripheries (Figure 5d).

3.2. Diagnosis of Obstacle Factors for RECC Enhancement and Their Spatial Characteristics

3.2.1. Identification of Dominant Obstacle Factors Across the Study Area

Based on the calculated obstacle degrees from 1331 valid sampling points, the factors restricting RECC enhancement in Daye City exhibit a hierarchical differentiation pattern: “dominated by mining-induced stress, strongly restricted by secondary ecological risks, and weakly constrained by agricultural baseline conditions” (

Table 4. Diagnostic results of the primary obstacle factors for RECC in Daye City.

Rank	Obstacle Factor	Subsystem	Obstacle Degree (%)
1	Spatial stress of industrial/mining land	Socio-economic	36.47
2	Geological hazard risk	Ecological environment	15.45
3	Soil erosion sensitivity	Ecological environment	13.01
4	Population density	Socio-economic	12.97
5	Location advantage index	Socio-economic	12.32
6	Soil organic matter content	Agricultural production	6.72
7	Arable land slope	Agricultural production	3.06

Note: A Friedman rank-sum test indicated significant differences among obstacle factors ($\chi$² = 6087.65, df = 6, p < 0.001). Kendall’s coefficient of concordance (W = 0.755) suggested strong agreement in factor rankings across spatial units.

). The top three primary obstacle factors are the spatial stress of industrial/mining land (36.47%), geological hazard risk (15.45%), and soil erosion sensitivity (13.01%). Together, they account for a cumulative obstacle degree of 64.93%, forming the core constraint chain that limits local RECC enhancement. Prolonged extensive mining in Daye City has generated contiguous abandoned industrial/mining land and associated ecological degradation, which is directly reflected in geological hazard and soil erosion risks, resulting in multiple interacting ecological constraints. In terms of socio-economic and agricultural production dimensions, the population density (12.97%) and location advantage index (12.32%) have a balanced share of the overall obstacle degree. This reflects the socio-economic imbalance in Daye City: a strong concentration of population and development resources in the urban core and relatively lagging development in peripheral rural areas. Furthermore, the combined obstacle degree for soil organic matter content (6.72%) and arable land slope (3.06%) is below 10%, indicating that the study area possesses relatively favorable agricultural soil baseline conditions, which impose only weak constraints on the overall regional RECC. To statistically verify the differences among obstacle factors, a Friedman rank-sum test was conducted using obstacle-degree values from all 1331 sampling units. The results showed significant differences among the seven obstacle factors ($\chi$² = 6087.65, df = 6, p < 0.001). Kendall’s coefficient of concordance (W = 0.755) further indicated strong consistency in factor rankings across spatial units, supporting the robustness of the identified obstacle hierarchy.

3.2.2. Spatial Agglomeration Patterns of Core Obstacle Factors

To further elucidate their spatial characteristics, the global Moran’s I was applied to test for spatial autocorrelation. The results indicate that the spatial stress of industrial/mining land (Moran’s I = 0.713), geological hazard risk (Moran’s I = 0.758), and soil erosion sensitivity (Moran’s I = 0.319) all exhibited significant positive spatial autocorrelation (all p < 0.001), indicating pronounced spatial clustering (

Table 5. Global Moran’s I statistics of the three core obstacle factors.

Factor	Moran’s I	Expected I	Variance	Z-Score	p-Value
Spatial stress of industrial/mining land	0.713	−0.0007	0.000173	54.20	<0.001
Geological hazard risk	0.758	−0.0007	0.000174	57.59	<0.001
Soil erosion sensitivity	0.319	−0.0007	0.000174	24.25	<0.001

). This confirms strong spatial clustering among the three primary obstacle factors, providing a robust statistical foundation for subsequent local spatial analyses. Subsequently, the Getis–Ord Gi* statistic was employed to map the spatial agglomeration patterns of these factors using an inverse-distance spatial weighting scheme and a fixed distance band of 2000.2 m (Figure 6). The analysis reveals that the highly significant hotspots (Z > 2.58, p < 0.01) of the three factors show substantial spatial overlap. They are predominantly concentrated in traditional core mining districts in north-central Daye City, including Jinhu Subdistrict, Chengui Town, and Jinshandian Town. This indicates that these districts are subject to multiple interacting constraints, underscoring their priority as primary governance units for regional spatial optimization and ecological restoration. Conversely, highly significant coldspots (Z < −2.58, p < 0.01) are primarily located across water bodies such as Bao’an Lake in the north and the forested Mufu Mountains in the south. Unaffected by significant mining disturbances, these areas serve as the ecological foundation of the study region. Furthermore, the majority of the city remains statistically non-significant (gray areas in Figure 6). This implies that mining-induced stress and its associated ecological risks are largely localized within specific mining districts and have not yet escalated into a systemic, region-wide ecological crisis. Consequently, there remains substantial buffer capacity for revitalizing existing land stock and implementing differentiated ecological restoration strategies.

3.2.3. Interaction Mechanisms Among Obstacle Factors

Prior to GeoDetector analysis, continuous variables were discretized into five categories using the K-means clustering method to maximize between-group differences. Single-factor analysis revealed that the three primary obstacle factors—spatial stress of industrial/mining land, geological hazard risk, and soil erosion sensitivity—all exhibited significant explanatory power for the spatial differentiation of RECC in Daye City (999 permutation tests, all p = 0.001). Geological hazard risk showed the highest explanatory power (q = 0.2338), followed by industrial/mining land stress (q = 0.1205) and soil erosion sensitivity (q = 0.1097). The discrepancy between the single-factor explanatory power and the obstacle degree rankings stems from the complementary analytical logic of the two methods. The obstacle degree model identifies common bottlenecks for regional RECC improvement, whereas GeoDetector reveals the driving mechanisms underlying RECC spatial differentiation by quantifying the explanatory power of individual factors. Together, the two approaches provide a multidimensional characterization of RECC constraints. Interaction analysis further revealed synergistic interaction effects among multiple factors (

Table 6. Interaction detection results of core obstacle factors in Daye City.

Interacting Factor Pair	q-Statistic (X1)	q-Statistic (X2)	Interaction q-Statistic	Interaction Type
Spatial stress of industrial/mining land ∩ Geological hazard risk	0.1205	0.2338	0.2823	Bivariate enhancement
Spatial stress of industrial/mining land ∩ Soil erosion sensitivity	0.1205	0.1097	0.2438	Nonlinear enhancement
Geological hazard risk ∩ Soil erosion sensitivity	0.2338	0.1097	0.2711	Bivariate enhancement

). All three factor pairs exhibited significant enhancement effects, with no independent or antagonistic relationships observed. Specifically, the interactions between industrial/mining stress and geological hazard risk, as well as between geological hazard risk and soil erosion sensitivity, demonstrated bivariate enhancement. Their interaction q-values (0.2823 and 0.2711, respectively) exceeded the explanatory power of either individual factor. Furthermore, the interaction between industrial/mining stress and soil erosion sensitivity exhibited nonlinear enhancement, with an interaction q-value (0.2438) greater than the sum of the explanatory power of the two individual factors. These findings indicate that RECC constraints in Daye City are not driven by isolated factors but by synergistically enhanced interactions rooted in long-term mining activities. Mining disturbances intensify geological hazard risks and accelerate soil erosion, triggering a cascading process of ecological degradation and declining carrying capacity. This compounding mechanism helps explain why RECC levels in the core mining districts are substantially lower than those in other parts of the city.

3.3. Optimized Layout of PLE Spaces

3.3.1. Identification and Classification of PLE Spaces Based on Land Use

Based on the functional attributes of land use types, PLE spaces in Daye City were identified and classified (Figure 7), and the area statistics are presented in Table 6.

Ecological space represents the dominant spatial category, accounting for 49.96% of the total area. It is primarily distributed in the southern mountainous regions and northern water bodies. Within this category, forest and grassland ecological space accounts for 35.92%, while aquatic ecological space accounts for 13.87%. Production space accounts for 36.07% of the total area and is dominated by agricultural production space (34.27%), which is mainly distributed in the flat areas of the central, northwestern, and southwestern regions. Mining production space accounts for 1.80%. Living space accounts for 13.97% of the total area and exhibits a combination of clustered and scattered patterns. Clustered areas are concentrated in the northern urban center, while scattered patches are distributed around towns and villages. Urban living space (7.71%) is slightly larger than rural living space (6.26%) and is mainly concentrated in the northeastern region.

3.3.2. Spatial Utilization Conflicts and Tailored Optimization Strategies

Building upon the quantitative diagnosis of obstacle factors and their spatial polarization patterns, the RECC-derived functional zoning was compared with the current PLE spatial classification in Daye City. This comparison identifies the typologies, formation mechanisms, and targeted optimization pathways of spatial conflicts (Figure 8). Conflicts in the study area fall into three categories: production–ecological, production–living, and ecological–living.

(1)

Production–Ecological Conflict: This conflict type largely overlaps with the high-value hotspots of soil erosion sensitivity—the third-ranked obstacle factor across the study area. As the most spatially extensive conflict, it covers 266.94 km² (47.73% of the total conflict area) and is concentrated in the transitional zone between the southern Mufu Mountains and the central plains (e.g., the peripheries of Liurenba and Yinzu Towns). Steep slopes and low vegetation cover render the area highly susceptible to soil erosion and rocky desertification triggered by steep-slope agricultural reclamation, thereby undermining regional water conservation functions. Optimization Strategy: In gently sloping areas with high-quality, contiguous cultivated land, terracing projects should be implemented. In steeply sloping areas with fragmented, low-quality land, the policy of cropland-to-forest and grassland conversion should be rigorously enforced to reinforce the city’s southern ecological barrier.

(2)

Production–Living Conflict: The primary driver of this conflict is the top-ranked obstacle factor—the spatial stress of industrial/mining land. Its highly significant hotspots largely coincide with structural conflicts in the central and western plains and gentle hills. Covering 222.74 km² (39.82% of the total conflict area), the conflict arises from two distinct mechanisms: (i) Urban expansion-driven conflict in the northeastern central urban area: The area has flat terrain and abundant water resources, affected by the expansion of the main urban area, and construction land has encroached on agricultural land. (ii) Structural conflict in the central and western plain and hilly areas: Inefficient rural residential land and industrial/mining land are fragmented and interspersed within large agricultural production areas, resulting in an imbalance in land allocation. Optimization Strategy: In the northeast, the urban–rural construction land linking quota policy should be applied to consolidate fragmented farmland while preserving the total cultivated land area, thereby reserving space for orderly urban expansion. In the central and western regions, underutilized construction land could be cleared and reclaimed as agricultural space to alleviate farmland fragmentation.

(3)

Ecological–Living Conflict: This conflict largely overlaps with the high-value hotspots of geological hazard risk—the second-ranked obstacle factor—further confirming the chain-reaction effect of secondary hazards induced by the spatial stress of industrial/mining land. It covers 66.95 km² (12.45% of the total conflict area) and appears as scattered patches in the southern low hills and piedmont zones. Some rural settlements are located within areas highly susceptible to landslides and debris flows, posing significant geohazard risks to local residents. Optimization Strategy: Systematic disaster risk assessments should be conducted for rural settlements, existing hazard sites remediated, and emergency facilities constructed. In high-risk areas, priority should be given to risk mitigation measures. Where appropriate, phased and voluntary relocation may be considered, followed by ecological restoration to enhance landscape stability.

3.3.3. Optimization of the Spatial Pattern

Drawing upon the three-dimensional RECC identification model, the evaluation scores of each subsystem were graded by importance and suitability. Adhering to the principle of “prioritizing ecological protection while ensuring production,” spatial zones with distinct functional orientations were delineated (Figure 9). The optimized spatial pattern is characterized as follows:

Ecological Functional Space (703.9461 km², 45.21%): This zone is predominantly distributed across the residual ranges of the Mufu, Longjiao, and Daji mountains in the south and east, together with northern lake wetlands such as Bao’an Lake and Daye Lake. It forms the ecological foundation of Daye City and performs core functions in water conservation and biodiversity conservation.

Agricultural Functional Space (521.7213 km², 33.51%): This zone is concentrated in the central and southwestern plains and gentle hills (e.g., Chengui, Lingxiang, and Jinniu towns). By consolidating fragmented arable land patches, the zone establishes a stable core hinterland for grain production across the study area.

Urban Development Space (331.3113 km², 21.27%): This zone is clustered in the northeastern urban core, encompassing Dongyue, Luoqiao, and Jinshan subdistricts as well as Huandiqiao Town to the north. It is designed to strengthen the agglomeration of population and industries, thereby reinforcing the region’s core urban development functions.

3.3.4. Effectiveness Evaluation of Spatial Pattern Optimization

To evaluate the effectiveness of territorial spatial restructuring, this study quantitatively compared the PLE spatial structure before and after optimization (

Table 7. Statistical table of the area of PLE spaces in Daye City.

Primary Category	Secondary Category	Area (km2)	Proportion (%)
Living space	Urban living space	120.0613	7.71
	Rural living space	97.5036	6.26
	Subtotal	217.5649	13.97
Production space	Agricultural production space	533.5440	34.27
	Mining production space	28.0447	1.80
	Subtotal	561.5887	36.07
Ecological space	Forest and grassland ecological space	559.3651	35.92
	Aquatic ecological space	215.9106	13.87
	Other ecological space	2.5499	0.16
	Subtotal	777.8257	49.96

and

Table 8. Comparison of PLE Spaces in Daye City Before and After Optimization.

Spatial Type	Proportion Before Optimization (%)	Proportion After Optimization (%)	Area Change Trend
Ecological space	49.96	45.21	Slight decrease (−4.75%)
Agricultural (production) space	34.27	33.51	Slight decrease (−0.76%)
Mining space	1.80	1.22	Slight decrease (−0.58%)
Living space	13.97	21.27	Coordinated expansion (+7.30%)

). The results indicate that Daye’s spatial development pattern has shifted from disorderly land expansion toward functional upgrading and higher-quality spatial development, as reflected in the following three aspects:

(1)

Mitigation of Core Obstacle Constraints: By identifying legacy and high-stress industrial and mining spaces, ecological restoration and land replacement were implemented for 8.96 km² of abandoned industrial and mining patches within the study area. These patches were systematically reallocated to ecological and agricultural spaces, thereby weakening the influence of industrial/mining stress (with an overall obstacle degree of 36.47%) on the regional RECC. Meanwhile, legally operating mines and certified green mines were retained to promote the transition of mining development from dispersed to intensive modes. This reduces spatial conflicts arising from fragmented industrial/mining land and supports the systematic remediation and functional optimization of inefficient industrial/mining land.

(2)

Resolution of Structural Spatial Conflicts: The total proportion of living space increased from 13.97% to 21.27%. This increase did not result from uncontrolled urban sprawl, but was achieved through the implementation of the linking policy for urban construction land increase and rural residential land decrease and disaster avoidance resettlement. These measures effectively alleviated the spatial interlacing conflicts between production–living spaces and ecological–living spaces.

(3)

Systematic Quality Improvement of Ecological and Agricultural Protection Spaces: The proportions of ecological space (45.21%) and agricultural space (33.51%) changed slightly after spatial optimization. The primary improvement resulted from the removal of misclassified degraded wasteland and polluted land patches. The optimized territorial pattern has clearer functional boundaries and higher spatial connectivity, thereby improving regional ecological resilience and strengthening the RECC baseline.

4. Discussion

This study reveals substantial differences in the constraint mechanisms of RECC between resource-exhausted cities and non-resource-based counties. Existing RECC studies on non-resource-based counties have mostly identified natural environmental conditions and climatic factors as the core drivers of spatial heterogeneity [50,51]. However, our findings indicate that the key constraint on RECC in Daye City, a typical mining-dependent city, is not its natural background but a chain reaction triggered by long-term mining activities. Spatial stress from industrial/mining land not only represents the dominant obstacle to RECC improvement but also exhibits significant synergistic enhancement with geological hazard risk and soil erosion sensitivity, ultimately forming a self-reinforcing process of ecological degradation and declining carrying capacity. This pattern is further supported by the pronounced spatial overlap of hotspot areas for the three factors in the core mining districts. Previous studies in Daye City and other mining-dependent regions have similarly reported that long-term mineral exploitation contributes to land degradation, ecological fragility, and elevated geological hazard risks [52,53], providing additional support for the constraint chain identified in this study. These findings help advance the quantitative understanding of mining-induced cascading effects on carrying capacity in resource-exhausted cities.

This study establishes an integrated analytical framework: carrying capacity assessment—obstacle diagnosis—spatial identification—targeted optimization, with clear practical and academic implications. In practice, the prioritized remediation units identified (e.g., Jinhu Subdistrict, Chengui Town) and the proposed multi-factor restoration strategy can inform mine ecological restoration and territorial spatial planning in Daye City. Academically, although Daye City has distinctive characteristics as a resource-exhausted mining city, the analytical framework proposed in this study is applicable to other mining-dependent regions facing similar challenges of resource exploitation, ecological degradation, and spatial development. By integrating RECC assessment, obstacle-factor identification, hotspot analysis, and GeoDetector, the framework provides a systematic approach for identifying key constraints and supporting differentiated spatial optimization. Nevertheless, the indicator system and management strategies should be adapted to local environmental, socio-economic, and planning conditions. This study also has certain limitations. Future research may incorporate long-term time-series data to further explore the dynamic coupling relationship between mining activities and the evolutionary trajectory of carrying capacity. Future studies could further evaluate the generalizability of the proposed framework through comparisons with other established assessment approaches.

5. Conclusions

Focusing on Daye City in Hubei Province, this study developed a RECC evaluation system from the three-dimensional perspective of ecological environment, agricultural production, and socio-economic development. Using the AHP–entropy weight combination method, obstacle degree model and other supplementary methods, we systematically examined the spatial differentiation characteristics of territorial space, the types of PLE space conflicts, and their corresponding optimization pathways for the study area. The core conclusions are as follows:

(1)

The RECC of Daye City exhibits a distinct north–south spatial differentiation pattern. High ecological carrying capacity is concentrated in the Mufu Mountains in the south and the lake-network region in the north, forming a spatial pattern shaped by mountain and water ecosystems. Cultivated land in the southern mountain fringe areas is subject to the dual constraints of soil erosion and geological hazards. High-value areas of agricultural carrying capacity are concentrated in the plains of the northern, central, and southwestern parts of the city, serving as the core hinterland for grain production. Urban development carrying capacity is highly coupled with the infrastructure density of the central urban area, exhibiting significant spatial polarization.

(2)

The spatial stress of industrial/mining land and its induced secondary ecological disasters constitute the core constraint chain for RECC improvement. The obstacle degree diagnosis shows that the spatial degree of industrial/mining land stress (36.47%) is the primary dominant obstacle to RECC enhancement. Together with geological hazard risk (15.45%) and soil erosion sensitivity (13.01%), these three factors reach a cumulative obstacle degree of 64.93%, jointly forming the core constraint system. Spatially, the extremely significant hotspots of the three factors exhibit a high degree of spatial overlap in the central and northern regions, dominated by legacy industrial/mining lands associated with historical resource extraction. Mechanistically, the interaction between industrial/mining stress and geological hazard risk yields a q-value of 0.2823, and its interaction with soil erosion sensitivity yields a q-value of 0.2438, presenting a strong bivariate enhancement and nonlinear enhancement, respectively. These results indicate a cascading feedback process linking mining disturbance, ecological degradation, and declining carrying capacity, which is the fundamental cause of the markedly low RECC in mining areas.

(3)

The spatial distribution of PLE space conflicts is highly coupled with the hotspot areas of core obstacle factors. Conflicts across the study area are dominated by production–ecological conflicts (accounting for 47.73%), which are accompanied by two core risks: ecological degradation and environmental hazards. Specifically, urban expansion-driven conflicts are concentrated along the urban fringe in the northeastern part of the city; structural conflicts are formed in the central and western rural areas under the influence of long-term industrial/mining activities; and the southern mountain fringe areas, constrained by both steep-slope farming and geological hazards, have significant safety risks for human settlements.

(4)

Differentiated spatial restructuring should focus on alleviating industrial and mining stress and mitigating cascading disaster risks. Based on the rigid constraints of RECC and obstacle diagnosis results, we constructed a territorial spatial optimization pattern characterized by “ecological barrier, contiguous agricultural land, and compact urban development”. After optimization, the proportions of ecological functional space, agricultural functional space, and urban development space are 45.21%, 33.51%, and 21.27%, respectively. For Daye City and other similar resource-exhausted cities, spatial restructuring should not be limited to land quota reallocation alone. For structural conflicts in the central and western regions, the focus should be on revitalizing existing land stock to improve land use efficiency; for high-risk areas in the south, strict ecological protection and risk-mitigation measures should be prioritized, while relocation may be considered where necessary and feasible. Meanwhile, a multi-stakeholder collaborative restoration and risk assessment mechanism could support improvements in human-settlement safety, resilience, and regional carrying capacity.