Policy-Driven Scenarios for Sustainable Peri-Urban Land Use: Production–Living–Ecological Space in Yubei District, Chongqing

Abstract

Sustainable land-use planning in peri-urban areas relies on informed decision-making guided by the examination of various development strategies. This study demonstrates a policy-based multi-scenario simulation which can serve as an aid to decision-making. Using the transformation of production–living–ecological (PLE) spaces in the Yubei District, a peri-urban district in Chongqing (2005 to 2020), as the baseline, the projections, simulated for 2035 under four scenarios, highlight the impacts of varying land-use policies: the reference scenario (RS), allowing unrestricted transformations, risks agricultural productivity and ecological integrity; the economic development scenario (S1) reveals the ecological costs associated with an economics-driven urban expansion; in contrast, the green development (S2) and agricultural land protection (S3) scenarios prioritize balanced growth and agricultural-land preservation so that ecological resilience and food security can be effectively maintained. Overall, significant land-use changes may occur, characterized by a substantial increase in living space, a decrease in production space, and stable ecological areas. This scenario-based analysis provides a comprehensive overview as to potential policy-driven planning outcomes. This aids in the identification of policy options that would best harmonize ecological, economic, and social objectives, offering essential insights for sustainable urbanization and land management in developing areas such as the Yubei District.

1. Introduction

The United Nations has adopted the 2030 Agenda for Sustainable Development, proposing 17 sustainable development goals to achieve the balance between the human socio-economic world and the natural environment [1]. Ecological civilization has become an ideological framework for China’s environmental conservation and an important concept in China’s processes aiming to achieve the sustainable development goals [2,3]. The proposal of an ecological civilization has promoted changes in land-use policies in China, especially in balancing production, living, and ecology [4]. In April 2015, the published opinions of the CPC Central Committee and the State Council on accelerating the construction of ecological civilization proposed that “land is the spatial carrier of the construction of ecological civilization. It is necessary to improve the spatial planning system, scientifically and reasonably lay out and renovate the production space, living space and ecological space [5]”. Deriving from this concept, the new territorial spatial planning (TSP) integrates previous main-functional-area planning, land-use planning, and urban and rural planning, and other related spatial planning efforts outlined in Chinese planning regulations and policies, into a unified spatial planning framework, one which is conducive to the alignment of the planning of economic and social development, environmental protection, urban planning, and land-use planning [6]. These efforts aim to push China further towards the ideal of sustainable land use.

In 2019, the Chinese government announced a plan to scientifically arrange production–living–ecological space (PLE space) to comprehensively address population distribution, economic layout, land use, and ecological protection [7]. Production space refers to areas focused on the creation of goods and materials including agriculture, industry, and related economic activities that support livelihoods [8,9,10]. Living space encompasses urban and rural areas utilized for residence, work, and everyday life [8,10,11]. Ecological space comprises artificial, semi-natural or natural vegetation; water bodies; and other ecological units that provide services essential to regional ecological security [12,13,14,15]. Establishing an ecological-space protection mechanism has also become central to China’s territorial spatial planning [7]. Rapid population growth and industrialization have expanded industrial and construction land, altering land cover and ecosystem structure. Ecological spaces predominantly overlap production and living space, and human activities have contributed to declining ecological quality [16]. The research on the conflicts among PLE spaces has become a crucial field for identifying pathways towards sustainable human development, with a particular emphasis on ecological spaces [17].

Such conflicts are especially severe in the regions where the population is still increasing and the economy is still growing, most particularly in some of the peri-urban areas in developing countries [18,19,20]. Studying land-use transition in these areas has become useful in evaluating the overall impact of anthropogenic activities on the nearby ecological system. As restricting or enforcing specific types of land use are often the means for the planner and the regulator to direct the development, the forecasted impact on the ecological system may help decision-makers avoid unintended impacts on the ecosystem.

With reference to China, the current research on ecological space predominantly focusses on identifying and classifying types of ecological space [21], studying the spatiotemporal dynamics [10], optimizing the configuration [22], and implementing zoning controls [23]. Landscape analysis [24], index analysis [25], spatial analysis [26,27], and modelling [17,28] are currently the primary methods employed in ecological-space research. These focus on ecological space, and the integration of these considerations into TSP is further illustrated by the establishment of the Ecological Civilization Construction Demonstration Area (ECCDA), which showcases the practical applications of these policies in promoting a balanced relationship between human activities and environmental protection.

In the context of ecological civilization, an ECCDA, designated by the Ministry of Ecology and Environment, is a national-level demonstration area that promotes harmony between humans and nature. Its development model is being replicated across China [29]. Ecological space, a core component of an ECCDA, enhances soil and water conservation capabilities, ecological services, and biodiversity maintenance, offering strong backing for ecological civilization efforts [17]. While central to an ECCDA, ecological-space production is also prioritized within TSP and within other related policies in China, having been a key characteristic of land-use management since the 2010s [6]. Land-use transitions in and around an ECCDA may reflect the dynamic and equilibrium between resource consumption and ecological support. The sustainability of such dynamics and equilibria signals the effectiveness of land-use decisions.

The Yubei District in Chongqing is one of ECCDAs in China (Figure 1). As one of the most dynamic development areas in Southwest China, Yubei has led all Chongqing areas, as to both permanent residents and GDP, since 2018 [30]. Driven by government development strategies, its sped-up socioeconomic development has led to the significant expansion of urban construction land to support industrial and economic development. These land-use changes require reasonable ecological-space planning to ensure sustainable local development and ecological civilization construction. As an ECCDA, Yubei faces national-level assessments and must balance rapid economic growth with ecological protection through effective spatial planning.

In this case, the contexts of the TSP and ECCDA, along with the local land-use and socio-economic development policies, have shaped or been affected by the past and present situations, collectively influencing the future situation of PLE space in the Yubei District. This case presents an opportunity to explore spatial-distribution characteristics under different development needs and the ecological effects of land-use transformation. Such understanding can inform future policies to optimize regional land-use patterns and align policy objectives, which is crucial for sustaining an ECCDA’s ecological space. Nevertheless, the current research exhibits two shortcomings: limited research on land-use transformation within ECCDAs, and inadequate simulation of district-level PLE spaces to assess ecological impacts. Hence, it is not easy to foresee whether the future trajectory of development in Yubei can be sustainable.

Policy-based scenario research is widely used in academia to evaluate policy effectiveness [31], assess urban development strategies for sustainability and livability [32], and evaluate energy transition policies and practices, as well as to understand the impacts of agricultural policies on food security and environmental sustainability [33]. This approach allows researchers and decision-makers to visualize potential futures, assess trade-offs, and identify best practices for implementing policies [34]. In this way, overly ambitious projects like Ordos and the Tianjin Eco-City may be avoided or improved [35,36,37]. Modeling land-use changes, which are key to both human activities and ecological processes, under the policy-based scenarios can provide an intuitive visual evaluation of the consequences of planning policies and environmental regulations. Recent research on policy-driven land-use scenario modeling have integrated policy objectives into spatial simulation frameworks [38,39,40]. The Markov–FLUS framework has demonstrated adaptability in diverse settings, effectively simulating scenarios based on policies related to economic development, ecological conservation, and farmland protection [41,42,43].

To address the research gaps and support planners and policymakers in advancing sustainable land development, this study demonstrates a new case-based simulation approach for PLE space, coupling ecological effect assessment. This approach models future spatiotemporal patterns of PLE space under different regulatory emphases and then evaluates the potential ecological effects. Selecting the Yubei District, a rapidly developing ECCDA area, as a case, this study simulates PLE changes in 2035 under four scenarios based on local and ECCDA-related policies. Further, this study demonstrates how ecological effects can be evaluated using the eco-environmental quality index derived from the modelled results. In this way, the study offers a practicable framework for forecasting land transition under different policy emphases, providing insights for advancing sustainable development, resilience strategies, and ecological civilization, not only in Yubei but in similar peri-urban areas undergoing land-use transformation.

2. Materials and Research Methods

In Section 2.1, we describe the geographical setting and socioeconomic context of the Yubei District, highlighting its dual pressures of urban development and ecological conservation. Section 2.2 then details all data sources and preprocessing steps, including land-cover datasets, administrative boundaries, and driving-factor rasters, along with data processing procedures. In Section 2.3, we present our methodological framework: first, introducing the integrated Markov–FLUS model and its validation metrics (Section 2.3.1); then outlining our land-use functional classification and eco-environmental quality index calculation (Section 2.3.2); and finally, we define policy-driven future scenarios for 2035 (Section 2.3.3). This structure ensures a logical progression from study area description, through data preparation, to modeling and scenario analysis.

2.1. Study Area

The Yubei District (106°27′30″–106°57′58″ E, 29°34′45″–30°07′22″ N) is located in the northeast of the urban area of Chongqing City (Figure 1). The level of economic development is currently ranked first among the districts in Chongqing, and hence the district is known as the main engine of Chongqing’s development [44]. As the vast area in the north of the Yubei District is located in Chongqing “Four Mountains” Nature Reserve (i.e., “Reserves of Jinyun Mountain, Zhongliang Mountain, Tongluo Mountain and Mingyue Mountain”, as shown in Figure 1) and adjacent to the Yangtze River, the balance between conservation and development has become the linchpin of the development of the Yubei District. In 2019, the Yubei District became the first national ECCDA in Chongqing, reflecting both its unique development status in Chongqing and the national concern for its future development and ecological conservation.

Additionally, the Yubei District is a peri-urban area [45], in which the agricultural population is rapidly transitioning to the non-agricultural sectors. The gradient of urban and rural landscape changes is sharp spatially, with an extensive and diverse industrial structure [46]. As a result, the ecological environment of the district experiences disturbances and is potentially subjected to stress, largely due to its high population density and intense social and economic activities. Confronted with the dual pressures of fostering ecological civilization and achieving high-quality regional economic development, the district must address increasingly stringent demands on the quality of its PLE spaces.

2.2. Data and Processing

The data used in this study mainly include land use/land cover data, vector boundary data and driving factor data. Land use/land cover data (30 m resolution) for 2010, 2015 and 2020 were obtained from the Data Centre for Resources and Environmental Sciences, Chinese Academy of Sciences (https://www.resdc.cn, accessed on 23 August 2024). Vector boundary data were obtained from the 1:1,000,000 national basic geographic database of the National Resource Catalogue Service for Geographic Information (www.webmap.cn, accessed on 23 August 2024), the Global Human Settlement Layer from the European Union (https://human-settlement.emergency.copernicus.eu/, accessed on 23 August 2024), and the Yubei District Planning and Natural Resources Bureau.

In terms of the driving factors of land use in Yubei, this study comprehensively considers the key drivers of urban development changes and the scale of the study area. Drawing on the extensive previous research literature, the driving factors selected for this study include following three main categories: natural environment, socio-spatial factors, and economic factors [47,48,49,50]. Prior to any analysis, all datasets were subjected to attribute and geometry-level quality control in ArcGIS 10.8. Spatial anomalies such as sliver polygons, overlapping boundaries, and topology errors were corrected via ArcGIS topology validation tools to ensure seamless layer integration. The data were then processed to ensure conformity with the requirements of the Markov–FLUS model by converting, projecting, and resampling the data into the same projection coordinate system with a spatial resolution of 30 m, the same resolution as the land use/land cover data.

We used altitude and slope as natural environment factors; GDP as an economic factor; and population, road network, and railway network, as well as points of interest (POI, i.e., in this research, town centers, airports, education spots, and shopping spots) as social–locational factors in the model. Altitude and slope were calculated from digital elevation model (DEM) data. GDP was derived from statistical yearbook data by using each town’s administrative center as a known GDP data point and applying inverse-distance weighting interpolation to generate an interpolated raster for the Yubei District [51]. Population raster data were derived from the Global Human Settlement Layer of the European Union, and the administrative boundaries of the Yubei District were applied to clip and unify coordinates and resolution. The distances to road and rail networks, and other POIs as social–locational factors, were calculated and normalized using the Euclidean distance and fuzzy membership degree tools of ArcGIS 10.8 [48]. Related data resources are detailed in Table 1.

Table 1. Data resources.

Data Types	Data	Data Sources
Land use/land cover data	Land use/land cover data for 2010, 2015, and 2020	Data Centre for Resources and Environmental Sciences, Chinese Academy of Sciences
Vector boundary data	Administrative boundary data	National Resource Catalogue Service for Geographic Information
	Natural reserve boundary data	Yubei District Planning and Natural Resources Bureau
	Permanent basic farmland boundary data	Yubei District Planning and Natural Resources Bureau
Driving factors data (Figure 2)	DEM	Yubei District Planning and Natural Resources Bureau
	Slope	Generated by the DEM data, using ArcGIS
	Point data (town center, airport, education spots, shopping spots)	National Resource Catalogue Service for Geographic Information
	Road data (railway, road)	National Resource Catalogue Service for Geographic Information
	Population	Global Human Settlement Layer
	GDP	Statistical Yearbook of Yubei District

Table 1. Data resources.

Data Types	Data	Data Sources
Land use/land cover data	Land use/land cover data for 2010, 2015, and 2020	Data Centre for Resources and Environmental Sciences, Chinese Academy of Sciences
Vector boundary data	Administrative boundary data	National Resource Catalogue Service for Geographic Information
	Natural reserve boundary data	Yubei District Planning and Natural Resources Bureau
	Permanent basic farmland boundary data	Yubei District Planning and Natural Resources Bureau
Driving factors data (Figure 2)	DEM	Yubei District Planning and Natural Resources Bureau
	Slope	Generated by the DEM data, using ArcGIS
	Point data (town center, airport, education spots, shopping spots)	National Resource Catalogue Service for Geographic Information
	Road data (railway, road)	National Resource Catalogue Service for Geographic Information
	Population	Global Human Settlement Layer
	GDP	Statistical Yearbook of Yubei District

Figure 2. Driving factors, suitability probabilities, and constraining factors of land use in Yubei: (A) DEM, (B) slope, (C) distance to town center, (D) distance to airport, (E) distance to educational sites, (F) distance to shopping spots, (G) distance to railway, (H) distance to road, (I) population, (J) GDP, (K) probability of suitability for land-use transformation, and (L) constraining factor.

Figure 2. Driving factors, suitability probabilities, and constraining factors of land use in Yubei: (A) DEM, (B) slope, (C) distance to town center, (D) distance to airport, (E) distance to educational sites, (F) distance to shopping spots, (G) distance to railway, (H) distance to road, (I) population, (J) GDP, (K) probability of suitability for land-use transformation, and (L) constraining factor.

2.3. Research Methodology

The methodology for this study integrates quantitative modelling and scenario analysis to simulate land-use changes and evaluate their ecological effects. It consists of utilizing the Markov–FLUS model to predict land-use changes based on historical data and driving factors, assessing model accuracy through metrics such as Kappa coefficient, OA, and the Figure of Merit (FoM), classifying land use into functional categories, calculating ecological quality indices, and designing multiple development scenarios to explore future land-use trajectories under varying socio-economic and environmental objectives.

2.3.1. The Markov–FLUS Model

The Markov–FLUS model represents an effective integration of the Markov chain model and the FLUS model which is used to forecast future changes in land types, and has demonstrated successful applications in various regions similar in scale to the region of the case study (i.e., the Yubei District), such as Xiong’an New Area Huo, Shi, Zhu, Xue and Chen [48], the Daxing District of Beijing [47], the Huangpi District of Wuhan [52], and the Asarsuyu watershed of Turkey [53]. The Markov chain model predicts the scales of future land change, based on current and historical data [54]. Subsequently, leveraging anticipated future land use scales, the FLUS model incorporates diverse driving factors influencing land type transitions, distributing spatial probabilities through a feedback mechanism integrating “top-down” and “bottom-up” approaches [55].

The FLUS model computes suitability probabilities for each land type using a neural network algorithm applied to initial land-cover data, coupled with factors impacting land-type transformations. Subsequently, employing cellular automata with an adaptive inertia mechanism and roulette algorithm, it simulates the spatial distribution of land types, considering developmental needs, suitability probabilities, cost matrices, neighborhood influences, and other pertinent conditions [47,48,50]. This study has been conducted following the principles related to the Markov–FLUS model detailed by Liu, Liang, Li, Xu, Ou, Chen, Li, Wang, and Pei [55].

Uncertainty in simulations is inevitable and arises from multiple sources, including the accuracy of the initial land-use data used, precision of driving factors (Figure 2A–J), and simulation performance [50,56]. To address these uncertainties, this study utilized historical land-use data from 2010 and 2015 to forecast land-use demand for 2020, and the actual land-use data in 2020 are compared to the simulated results to check the precision (Kappa coefficient, OA, and FoM in this study) of the model (Figure 2). Data were integrated concerning suitability probabilities (Figure 2K) and constraining factors (Figure 2L) specific to each land-use type, with the Markov model applied to project changes in land use within the Yubei District from 2015 to 2020.

To assess the accuracy of the Markov–FLUS model, the Kappa coefficient is employed as a measure representing the proportion of correct categorization, used to describe the reduction in errors compared to completely random categorization. The overall accuracy (OA) represents the general agreement between the reference map and the simulated one, typically expressed as a percentage. A 100% accuracy signifies a perfect classification, in which all reference sites were accurately classified [57]. Both OA and the Kappa coefficient are utilized to verify the validity of the Markov–FLUS model, with values closer to one indicating higher confidence in the simulation. When both OA and Kappa are equal to or greater than 0.7, it is generally considered that the model’s simulation demonstrates a high degree of consistency with the actual situation [58]. Additionally, this study employs the Figure of Merit (FoM), a coefficient recognized for its superior ability to characterize simulation accuracy relative to the Kappa statistic in order to assess model precision [59,60]. FoM values approaching 1 denote increasingly better simulation performance and greater accuracy. However, the empirical evidence indicates that FoM scores typically fall below 0.3 [59], with the 0.1–0.2 range being most common [55,61,62]. Parameters such as Kappa, OA, and FoM serve to verify the suitability and precision of the Markov–FLUS model in simulating the future spatial distribution of land use in the case area. Figure 3 presents a comparison between the predicted outcomes and actual 2020 land-use patterns. The sub-figures include selected areas surrounding water bodies and rivers, comparing the differences between actual and simulated land use at a smaller scale. This is because riversides can represent diverse types of land use, such as agricultural land, forest, and construction land. Therefore, locations near water systems often better reflect changes in land-use types [63,64]. The overall accuracy of the simulation in this research is 0.8480, the Kappa coefficient is 0.7537, and the FoM coefficient is 0.21, which resides comfortably within the expected interval, indicating acceptable simulation accuracy. The differences observed between the predicted and actual datasets are because of the accelerated urbanization in the Yubei District after the year of 2015, in which the growth rates of the living space and industrial and mining production space are higher than the historical trend. In Figure 3, the main differences between the simulated and actual 2020 land-use patterns occur in the towns of Longxing, Shichuan, and Luoqi in southeastern Yubei District (the locations of these towns are shown in Figure 1), in which the growth rates of living space and industrial and mining production space exceed historical trends. This difference is largely due to recent investments in the Longshi Industrial Park and the Luoqi Port and Logistics City, which have driven significant policy-related increases in living space and industrial and mining production space, causing the FLUS model to struggle to capture potential changes accurately. In the 2035 land-use simulation, the baseline data have been updated to match the actual 2020 land-use pattern, and policy factors have been incorporated into the simulation framework, effectively addressing this limitation.

2.3.2. Land-Use Classification, Land-Use Change, and Evaluation of Ecological Effects

One type of land use normally has a dominantly leading function within the aspects of production, living, and ecology, despite their multifaceted functions [65]. According with the national land-use classification standard in China (GB/T21010-2017) [66] and the research of Li and Fang [65] and Liu, Liang, Li, Xu, Ou, Chen, Li, Wang, and Pei [55], this study considers the functions of land-use types and divides land types into eight land-use types, the seven of which are classified as production space, living space or ecological space [55] (

Table 2. Land-use function classification of PLE space, and eco-environmental quality index.

Classification of PLE Space		Secondary Classification of Land-Use Classification System	Eco-Environmental Quality Index
1st Class	2nd Class
Production space	Agricultural production space	Arid land, Paddy land	0.2514
	Industrial and mining production space	Other construction land	0.1500
Living space	Urban and rural living space	Urban built-up land, Rural settlement	0.2000
Ecological space	Forest ecological space	Woodland, Shrub land, Spare woodland, Other woodland	0.7937
	Meadow ecological space	High coverage grassland, Medium coverage grassland, Low coverage grassland	0.6220
	Water ecological space	River canal, Reservoir pond	0.5500
	Other ecological space	Beach land, Swamp	0.5530

).

Land-use change analysis is a way to study how human activities, and natural processes, shape the earth’s surface over time. By quantifying changes in land cover and use, this method helps track trends like urban growth, deforestation, and the spread of agriculture. Land-use change analysis can be used for managing natural resources, understanding environmental impacts, and making better decisions about land-use planning and policies [67,68]. Building on the understanding of land-use changes, the eco-environmental quality index provides a tool for assessing the ecological impacts of land-use shifts.

The eco-environmental quality index is used to evaluate the changes in ecological effects. Previous research has utilized expert scoring methods to calculate the eco-environmental quality index for various land-use types, which were then normalized within the [0, 1] range [69,70,71]. This index effectively reflects the ecosystem status in China and has gained widespread adoption [72,73,74].

Using this index for each subcategory of land-use function classification, the eco-environmental quality index of the PLE space in the Yubei District was computed using a weighting method (Table 2). The indices of eco-environmental quality associated with various land-use types exhibit variations across the research areas, and changes in the index are mainly attributed to shifts in both the total area and the land-use area of each land-use type in the region. This can be articulated as follows:

$${\mathrm{E}\mathrm{Q}}_{\mathrm{t}}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}\frac{{\mathrm{A}}_{\mathrm{k}\mathrm{i}}}{{\mathrm{A}}_{\mathrm{k}}}{\mathrm{V}}_{\mathrm{i}}$$

(1)

where ${\mathrm{E}\mathrm{Q}}_{\mathrm{t}}$ is the eco-environmental quality index of Yubei in the period t; n is the number of land-use types in the area; ${\mathrm{A}}_{\mathrm{k}\mathrm{i}}$ is the area of land type i in the k ecological unit (the total area of Yubei in this study); ${\mathrm{A}}_{\mathrm{k}}$ is the total area of the ecological unit k; and ${\mathrm{V}}_{\mathrm{i}}$ is the value of the eco-environmental quality index of the i land-use type.

2.3.3. Scenario Settings Under Development Needs and Goals

The co-evolution of the socio-economic and natural environments is marked by uncertainty [50]. Scenario analysis describes and compares the conditions of various scenarios, projecting future conditions that represent developmental objectives, based on diverse assumptions. This approach considers a spectrum of potential future circumstances to inform the development of strategies best suited for future planning [75,76]. Policy scenarios comprise structured narratives or models that describe potential future conditions based on varying policy choices and external factors [77]. They serve as a tool for policymakers to visualize the implications of different strategic decisions, assess uncertainties, and evaluate potential outcomes within complex systems [78]. This study developed four development scenarios of land use in the Yubei District in 2035 based on the national ECCDA standards and various development policies and planning data released by Chongqing City and the Yubei District (

Table 3. Policy bases for the proposed scenarios.

Policy Number	Policy Name	Policy Issuing Authority	Source of Policy Content
P1	Management Regulations for an ECCDA	Ministry of Ecology and Environment of the People’s Republic of China	https://www.mee.gov.cn, accessed on 25 August 2024
P2	Construction Indicators for an ECCDA	Ministry of Ecology and Environment of the People’s Republic of China	https://www.mee.gov.cn, accessed on 25 August 2024
P3	TSP of Chongqing (2021–2035)	Chongqing Municipal Bureau of Planning and Natural Resources	https://ghzrzyj.cq.gov.cn, accessed on 25 August 2024
P4	Outline of the 14th Five-Year Plan for National Economic and Social Development and Long Range Objectives for 2035 for the Yubei District	Yubei District People’s Government of Chongqing Municipality	http://www.ybq.gov.cn, accessed on 25 August 2024
P5	TSP of the Yubei District (2021–2035)	Yubei District People’s Government of Chongqing Municipality	http://www.ybq.gov.cn, accessed on 25 August 2024
P6	14th Five-Year Plan for Ecological Environmental Protection in the Yubei District	Yubei District Development and Reform Commission	http://www.ybq.gov.cn, accessed on 25 August 2024

).

The four scenarios set in this study (hereinafter labeled as RS, S1, S2, and S3) are briefly described in the following:

The reference scenario (RS), termed the natural development scenario, is formulated based on the historical trajectory of economic, population-based, and technological advancements in the Yubei District. This scenario assumes the sustained continuity of these trends, along with the land-use changes usually driven by these socio-economic developments. To reflect this assumed continuation of land-use change patterns, the transition probabilities are maintained at levels observed between 2015 and 2020. These transition probabilities for land use in this scenario also serve as a reference for adjusting the transition probabilities in the remaining scenarios (S1, S2, and S3), where changes are made based on specific policy emphases. Depending on the policy priorities, the transition probabilities for relevant land use are increased or decreased from this baseline.

The economic development scenario (S1) is set to maximize socio-economic benefits, and assumes that the city becomes increasingly attractive for immigration due to rapid regional economic growth and technological innovation. In this scenario, economic development is given priority, and rapid expansions of construction land such as cities, roads, and industrial and mining land use, which are important symbols of economic development, are necessary. The cost of conversion from construction land to other types of land is set to be the highest among S1, S2, and S3, and the probability of transferring from living space to production space and ecological space is set as reduced. The formulation of this scenario is mainly based on local (Chongqing and Yubei District) land-use policies, which have made plans and considered prospects for the future economic development of the region (P3, P4, and P5, listed in Table 3). In accordance with the TSP-based regulation that the expansion ratio of urban development boundaries should be controlled within 1.3 times that of 2020 (P5), the area of construction land in this scenario is limited to 1.3 times the area of construction land in 2020. Given the complexity of policy implementation, our calculations demonstrate that, under the existing farmland-protection and ecological-red line policies, the theoretical maximum expansion of construction land can be greater than 1.3 times of its current level. Consequently, the design of this scenario still satisfies the ecological and farmland protection red lines stipulated in the current TSP in the Yubei District.

The green development scenario (S2) prioritizes the protection of the ecosystem. Thus, the transition probabilities of ecological-space land are reduced to the lowest among all scenarios, enabling the expansion of ecological space, or at minimum, significantly slowing down the shrinkage of ecological space. This setup reflects the emphasis on strengthening the protection and maintenance of forests, grassland, water bodies, and other ecological lands, while restricting the expansion capacity of the other land types relative to these areas. Specifically, according to the regulations for an ECCDA and the relevant visions for the Yubei District (namely P1, P2, P4, and P6, as listed in Table 3), the ecological protection red line must be protected, the forest coverage rate should reach 55%, and the district’s ecological function should not fall below the corresponding level in 2020. For the simulation, therefore, the transition probabilities for ecological-space land use are set at the lowest among all the scenarios, while the transition probabilities for converting production-space land into ecological-space land are increased. Reflecting the practical constraints, the transition probabilities of construction land reverting to the condition of ecological space remain low. Collectively, however, these parameter settings aim to maintain or improve the forest coverage rate, ensuring that it remains at 55% or above.

The cultivated-land protection scenario (S3) is designed to consider the impact of cropland protection policies. The main objective of these policies is to effectively safeguard cropland resources and slow down the shrinkage of production space. Accordingly, in the simulation setup, the cost of converting cultivated land into other land-use types (including other production-space lands or ecological-space lands) is increased, while the cost of converting other land types to cropland is reduced, except for built-up land. The policy basis for this scenario draws from P3, P4, and P5 (Table 3), which adheres to the strictest farmland protection system and ensures that by 2035, the farmland area in the Yubei District will not be less than 209.88 km². This minimal farmland-area target provides the numerical foundation for adjusting the transition probabilities between all land-use types in the simulation.

In addition, the conversion cost matrix is represented by 0 and 1. In cases where one land type is prohibited from converting to another, the corresponding entry in the matrix is set to 0; conversely, if the conversion is permissible, it is set to 1. The neighborhood factor parameter signifies the ease of transition from one land class to another, ranging between 0 and 1. A higher value indicates a greater propensity for expansion in the land-use type. The estimation of neighborhood factors is based on an analysis of historical land-use data within the study area, supplemented by expert judgment [41,61]. Drawing from both simulation outcomes and the specific context of the Yubei District, the conversion cost matrix (

Table 4. Conversion cost matrix.

Scenarios	PLE Space Type	A	B	C	D	E	F	G
Reference Scenario (RS)	A	1	1	1	1	1	1	1
	B	1	1	1	1	1	1	1
	C	1	1	1	1	1	1	1
	D	1	1	1	1	1	1	1
	E	1	1	1	1	1	1	1
	F	1	1	1	1	1	1	1
	G	1	1	1	1	1	1	1
Economic Development (S1)	A	1	0	1	1	1	1	1
	B	1	1	1	1	1	1	1
	C	1	1	1	1	1	1	1
	D	1	0	1	1	1	1	1
	E	1	0	0	1	1	1	1
	F	1	0	0	1	1	1	1
	G	1	0	0	1	1	1	1
Green Development (S2)	A	1	0	1	0	0	0	1
	B	1	1	1	1	0	0	0
	C	1	1	1	1	0	0	0
	D	1	0	1	1	1	0	1
	E	1	0	0	0	1	0	1
	F	1	0	0	0	0	1	1
	G	0	0	0	0	1	0	1
Cultivated Land Protection (S3)	A	1	1	1	1	1	1	1
	B	1	1	1	1	1	1	1
	C	1	1	1	1	1	1	1
	D	1	0	1	1	1	1	1
	E	0	0	0	0	1	1	1
	F	0	0	0	0	1	1	1
	G	0	0	0	0	1	1	1

(In terms of the PLE space type, A refers to agricultural production space, B refers to industrial and mining production space, C refers to urban and rural living space, D refers to forest ecological space, E refers to meadow ecological space, F refers to water ecological space, and G refers to other ecological spaces).

) and neighborhood factor parameters (

Table 5. Neighborhood factor parameters.

Scenarios	Agricultural Production Space	Industrial and Mining Production Space	Urban and Rural Living Space	Forest Ecological Space	Meadow Ecological Space	Water Ecological Space	Other Ecological Space
Reference Scenario (RS)	0.5	1	1	0.7	0.3	0.4	0.01
Economic Development (S1)	0.2	1	1	0.5	0.2	0.4	0.01
Green Development (S2)	0.5	0.7	0.7	1	0.7	0.3	0.01
Cultivated Land Protection (S3)	0.8	0.8	0.8	0.7	0.3	0.3	0.01

) for the proposed scenarios were established, leveraging insights from previous research [79,80,81] through multiple experiments.

3. Results

3.1. Changes in Production–Living–Ecological Spaces from 2005 to 2020

From 2005 to 2020, the land-use change analysis of the PLE space in the Yubei District revealed a trend characterized by “one increase, one decrease, and one stability” (Figure 4). The living space in 2020 increased to more than three times the corresponding space in 2005. The production space in 2020 shrunk slightly, more than 10%, in comparison to that of 2005. Notably, as an ECCDA, the ecological space of the Yubei District has remained stable or slightly improved throughout the 15 years of the study, in line with the national policy requirements for an ECCDA. From the perspective of change amplitude, the total change in PLE spaces in the Yubei District from 2015 to 2020 was the most significant, compared to the previous two five-year periods, mainly due to the transformation of a large amount of production space into living space.

The detailed mutual transformation between production, living and ecology spaces and their spatial distribution are further explored. As shown in Figure 5, from 2005 to 2020, 140.33 km² of the production space of Yubei was converted into other types of space. The area is the largest amount of land transferred out among all space types, most of which was transformed into living space (119.57 km²) (this type of space received the most land from other types of space). Relatively little land-transfer happened from living space to production space (3.53 km²), or from living space to ecological space (0.43 km²). In terms of ecological space, 12.13 km² has been converted into production space, and another 3.64 km² has been converted into living space. It is clear that the land-use transformation in the Yubei District is mainly achieved through the transition from production space to living space. Ecological space, both in size and distribution, underwent minimal land-use changes with respect to production and living spaces, and was dynamically replenished to maintain stability1.

This study further explores the characteristics of the transfer of production space to living space, which contributed the largest portion of land-use transition, by analyzing the land-use transfers under second-class categories (Table 2). The analysis revealed that the transfer of agricultural production space to urban living space (112.31 km²) contributed most to the conversion from production space to living space (Figure 6). In the past 15 years, the industrial and mining production space has increased by 125.71 km². Within this space, 123.34 km² were obtained through the transfer of agricultural production space. Within the category of living space, little change has occurred in rural living space (a decrease of 1.89 km²), while a disproportionally large expansion of urban living space (121.13 km²) has been observed. Overall, a large amount of agricultural production space in the Yubei District has been transformed into urban living space and industrial and mining production space.

3.2. Multi-Scenario Simulation Results

On the whole, the multi-scenario simulation describing PLE space in the Yubei District in 2035 showed that with specific scenarios setting targets for proposed development needs, each scenario produced quite a different PLC space distribution. Taking the development situation in 2020 as the baseline, the amounts of ecological space under the RS and S1 scenario decrease, while the amounts of ecological space under the S2 and S3 scenario rise. The shrinking of agricultural production space has contributed to the overall decrease in production space in all scenarios, with the lowest decrease observed in S3. By contrast, industrial and mining production space has increased slightly in all scenarios. Living space has increased in all of the various scenarios (

Table 6. Land-use changes in PLE space under RS.

	Production Space		Living Space	Ecological Space
	Agricultural Production Space	Industrial and Mining Production Space		Forest Ecological Space	Meadow Ecological Space	Water Ecological Space	Other Ecological Space
Land use RS 2035	775.14	138.59	234.65	249.78	36.54	21.96	0.41
Land use 2020	825.38	134.38	173.61	260.98	37.21	25.12	0.37
Land-use change	−50.24	4.21	61.04	−11.21	−0.67	−3.16	0.04
Change rate	−3.66%	3.13%	35.16%	−4.29%	−1.80%	−12.58%	10.81%
PLE RS 2035	913.73		234.65	308.69
PLE 2020	959.77		173.61	323.68
PLE change	−46.04		61.04	−15.00
Change rate	−4.80%		35.16%	−4.63%

).

From a spatial-distribution perspective (Figure 7), the living space and non-agricultural production space (industrial and mining land) in the Yubei District mainly expand slightly from southwest to northeast. Living spaces and non-agricultural production spaces have been added to the mountainous plains in the southeastern part of the Yubei District and the agricultural land surrounding the original city. The overall trend of agricultural production space in the Yubei District is decreasing, and the spatial distribution of the reduced agricultural production space in various scenarios is basically consistent with the spatial-distribution characteristics of the increased living space and non-agricultural production space.

From the perspective of geomorphic units, the relatively extensive ecological space is mainly distributed around the three mountain ranges and several water bodies in the Yubei District. Across all scenarios, changes in ecological space are mostly confined to areas adjacent to the 2020 baseline, reflecting the strength of the model in maintaining the spatial continuity during the future-land-use simulation. The clear gradient transitions of living space, production space and ecological space helped maintain the integrity of the ecological space, preventing serious fragmentation by built-up areas. In our simulation results, the production space, especially agricultural land, serves as a buffer between the ecological space and living space.

3.2.1. RS: The Reference Scenario

The reference scenario pertains to unrestricted transformations in land use, in which such changes are predominantly shaped by the competing influences of the natural environment and the social and long-term economic development trends within the study area, without any limitations being imposed by land development policies. Consequently, there are no restrictions on land transfer, facilitating the unrestricted conversion of different land-use categories in this simulation. Figure 7B shows the simulation of land-use distribution in 2035 in this scenario, while Table 6 presents the changes in land-use space areas from 2020 to 2035.

In comparison to 2020, the extents of the agricultural production space, forest ecological space, meadow ecological space, and water ecological space are projected to diminish by 2035, with reductions of −3.66%, −4.29%, −1.80%, and −12.58%, respectively. Conversely, both industrial and mining production space, as well as living space, are anticipated to expand, with the latter experiencing a notable increase, specifically, a growth of 35.16% (234.65 km²). In the reference scenario, the rapid expansion of construction land, primarily driven by human activities to satisfy the demands of socio-economic development, has led to the conversion of agricultural production space and ecological spaces.

Spatially, the expansion of living space is an internal expansion in the southwest of the Yubei District on the basis of the original distribution, and extends slightly into the northeast, occupying the original agricultural production space, with a relatively concentrated distribution (Figure 7A,B). This is consistent with the changing trend from 2005 to 2020 associated with rapid urbanization. The findings from this scenario suggest that the rapid development of socio-economic conditions in the Yubei District in the future will lead to urbanization, resulting in further expansions in the scale of construction land, mainly focused on providing more living spaces.

This trend is accompanied by corresponding declines in both agricultural production space and forest ecological space. The phenomenon of urban development into agricultural land resources is particularly severe, and the expansion of urban areas has consequently resulted in a reduction of ecological spaces.

The modelling results show that within the RS, unconstrained growth will lead to a swift increase in regional construction land, accompanied by significant declines in agricultural production and ecological areas. This phenomenon will hinder the region’s ability to achieve development balanced among its ecological, social, and economic systems.

3.2.2. S1: The Economic Development Scenario

The economic development scenario incorporates the actual situation in the Yubei District, which is undergoing a phase of rapid economic development, as well as regional land-use policy. In this scenario, the transfer probability associated with construction land (including industrial and mining production space as well as living space) relative to other land-use spaces is reduced from the baseline values used in RS.

The simulated trends and spatial differences in the changes in among land-use types in 2035 in S1 (Figure 7A,C and

Table 7. Land-use changes in PLE space under S1.

	Production Space		Living Space	Ecological Space
	Agricultural Production Space	Industrial and Mining Production Space		Forest Ecological Space	Meadow Ecological Space	Water Ecological Space	Other Ecological Space
Land use RS 2035	778.73	174.70	193.20	248.01	37.01	25.04	0.37
Land use 2020	825.38	134.38	173.61	260.98	37.21	25.12	0.37
Land-use change	−46.65	40.32	19.59	−12.98	−0.20	−0.08	0.00
Change rate	−5.65%	30.00%	11.28%	−4.97%	−0.54%	−0.32%	0.00%
PLE RS 2035	953.43		193.20	310.43
PLE 2020	959.77		173.61	323.68
PLE change	−6.34		19.59	−13.26
Change rate	−0.66%		11.28%	−4.09%

) are generally like those in RS. Agricultural production space, forest ecological space, meadow ecological space, and water ecological space decrease in area, while the areas of industrial and mining production space and living space increase dramatically. Despite the similarity, the growth rate of industrial and mining production space in S1 is clearly higher than that in RS, increasing from 134.38 km² in 2020 to 174.70 km² in 2035, with a growth rate of 30.00%, in comparison to 3.13% in RS. The trend of decreasing agricultural production space is more severe, from 825.38 km² in 2020 to 778.73 km² in 2035. In addition, most ecological-space land-use types, excluding other ecological spaces (i.e., beach land, swamp), show a decreasing trend, indicating that under this scenario, the expansion of urban areas leads to a reduction in the size of lands associated with the ecological land-use type. This could result in a decline in regional ecological sustainability.

Notably, the growth in the area of construction land in this scenario is 1.19 times that of 2020, which is in line with the TSP regulations of the Yubei District stating that “the expansion ratio of urban development boundaries should be controlled within 1.3 times that of 2020” (P5, Section 2.3.3). Despite reduced agricultural and ecological spaces, the simulation of this scenario meets the policy expectations of TSP. However, there is a requirement in P2 that potentially cannot be met in this scenario, namely, a relatively qualitative description: “Maintaining stability or continuous improvement of ecosystems.” Although the language of the policy is somewhat ambiguous, the economic development strategy appears to conflict with the principle declared in the ECCDA framework in the context of ecological civilization. This is an issue that should be addressed by policymakers.

3.2.3. S2: The Green Development Scenario

In alignment with various strategies implemented by the Chinese government regarding the context of ecological civilization, this research formulates the green development scenario (S2). To safeguard regional ecological integrity, it is imperative to rigorously protect areas that significantly influence the ecological environment, including forests, grasslands, and water areas, while prohibiting extensive development and exploitation in these regions, as outlined in land development policy for the Yubei District (Table 3). The ecological conservation area is designated as a restricted conversion zone in this scenario and the probability of converting any ecological-space land types to other types has been reduced.

Compared with 2020, the forest ecological space, meadow ecological space, and water ecological space show a slight increase in 2035, with growth rates of 9.47%, 0.16%, and 0.16%, respectively (

Table 8. Land-use changes in PLE space under S2.

	Production Space		Living Space	Ecological Space
	Agricultural Production Space	Industrial and Mining Production Space		Forest Ecological Space	Meadow Ecological Space	Water Ecological Space	Other Ecological Space
Land use S2 2035	771.86	144.37	192.33	285.70	37.27	25.16	0.37
Land use 2020	825.38	134.38	173.61	260.98	37.21	25.12	0.37
Land-use change	−53.53	9.99	18.72	24.72	0.06	0.04	0.00
Change rate	−6.48%	7.43%	10.78%	9.47%	0.16%	0.16%	0.00%
PLE S2 2035	916.23		192.33	348.50
PLE 2020	959.77		173.61	323.68
PLE change	−43.54		18.72	24.82
Change rate	−4.54%		10.78%	7.67%

). The changes in land-use spaces in this scenario still primarily concentrate on agricultural production space, industrial and mining production space, and living space. Agricultural production space sees its area further compressed by −6.48%, to only 771.86 km².

The expansion of living space is evident (10.78%, Table 8), but its expansion rate is more controlled than in RS (36.16%, Table 6) and S1 (11.28%, Table 7). Compared to S1, the increase in the amount of industrial and mining production space has also decreased from 30.00% to 7.43%. It is expected that this scenario would satisfy the needs for urban economic and social growth by intensifying urban development rather than expanding; the built-up areas in Yubei will need to be used more effectively.

Overall, to uphold the stability of ecological spaces and accommodate socio-economic activities, the predominant trajectory of agricultural-land conversion continues to be toward construction land, encompassing industrial and mining space as well as living space. Particularly, within the green development scenario, ecological space—including forests, grasslands, and water bodies—demonstrates a trend of growth, thereby positioning agricultural land as the principal category undergoing conversion. The decrease in the encroachment of construction land into ecological areas could preserve regional ecological security.

3.2.4. S3: The Agricultural-Land Protection Scenario

Considering that agricultural-land protection is essential for the effective conservation of farmland resources, except for basic farmland that cannot be converted into other land types, the cost of agricultural-land conversion in this scenario has been increased to restrict the transfer and change of agricultural land to other land types.

According to this scenario, the area of agricultural production space is 816.53 km², which is the largest value among the simulation results, with a decreasing trend of only −1.07% compared with 2020 (

Table 9. Land-use changes in PLE space under S3.

	Production Space		Living Space	Ecological Space
	Agricultural Production Space	Industrial and Mining Production Space		Forest Ecological Space	Meadow Ecological Space	Water Ecological Space	Other Ecological Space
Land use RS 2035	816.53	137.67	177.62	261.13	38.10	25.65	0.37
Land use 2020	825.38	134.38	173.61	260.98	37.21	25.12	0.37
Land-use change	−8.86	3.29	4.01	0.14	0.89	0.53	0.00
Change rate	−1.07%	2.45%	2.31%	0.06%	2.39%	2.11%	0.00%
PLE RS 2035	954.20		177.62	325.24
PLE 2020	959.77		173.61	323.68
PLE change	−5.57		4.01	1.56
Change rate	−0.58%		2.31%	0.48%

). Distinct from other scenarios, some high-quality cultivated land near the urban area of the Yubei District with relatively flat terrain has not been converted to other land-use types (Figure 7E).

Each type of ecological space has increased slightly, contributing to the increase in total ecological space of 0.48% compared to 2020, indicating that the protection of agricultural production space has a relatively small impact on ecological space. The growth rates of industrial and mining production space and living space in this scenario are the smallest, at 2.45% and 2.31%, respectively. This indicates that the rate of urban expansion will be somewhat regulated through the enforcement of agricultural-land protection measures.

In summary, the simulation results indicate that agricultural-land conversion has been effectively slowed down by implementing restrictive factors in basic farmland protection areas and increasing the cost of the land conversion rate. Accordingly, assuming that the policy to protect agricultural land is implemented, the quantity of agricultural land can be preserved. Despite these measures, the rapid development of the economy in the Yubei District will inevitably lead to the expansion of industrial and mining production space as well as living space. In addition, the competing policy goals related to ecological civilization aiming to protect the ecological space are expected to add pressure to protect the agricultural production space in the Yubei District.

3.3. Comparative Analyses of the Multi-Scenarios and Eco-Environmental Effect

Overall, the four scenarios reveal some common trends in PLE space changes: a reduction in production space, an increase in living space, and a relative stability in ecological space.

The most pronounced shift from production to living space occurs in the RS scenario (Figure 8A), highlighting that, under conditions of unrestricted natural growth, a substantial amount of agricultural land will be converted into new residential areas. In scenarios S1, S2, and S3, which consider policy implications with different focuses, the area of production space transitioning to living space decreases, particularly in the S3 scenario, while the constraints of cultivated-land protection result in only a minimal portion of agricultural production space being replaced by living space. The simulation results indicate that emphasizing different land-use policies significantly influences the spatial patterns of land use.

In the RS and S1 scenarios, which prioritize economic development, some ecological space is converted into production and living spaces. Conversely, in the green development scenario (S2), a portion of ecological space continues to transition into production and living uses; however, some production space is also converted back into ecological space, achieving a dynamic balance in ecological space. In the agricultural-land protection scenario (S3), the conversion of agricultural land is further restricted, and ecological space within protected areas is likewise unable to be transformed into production or living space, resulting in minimal transitions across all land-use spaces.

The eco-environment quality index (Equation (1)) for the Yubei District in 2035 is calculated for each scenario (

Table 10. Eco-environmental effects under the proposed scenarios.

Year/Scenario	Ecological Environment Quality
2020	0.3477
2035/RS	0.3417
2035/S1	0.3394
2035/S2	0.3556
2035/S3	0.3478

). In 2035, the eco-environmental quality index of the Yubei District is 0.3417, 0.3394, 0.3556, and 0.3478 in scenarios RS, R1, R2 and R3, respectively. Compared to 2020 (0.3477), the index increased in the two scenarios addressing ecological functions and some production functions, while it decreased in the two scenarios emphasizing economic development. The eco-environment quality index under S2 (the green development scenario) was the highest.

The calculation of the environmental quality reflects the general idea that the expansion of ecological space (with higher weights, see Table 2) can improve ecosystem services. Yet, the degree of improvement depends on the perceived benefit of the secondary land-use classification within the ecological space (Table 2). In S2 and S3, the increasing of ecological spaces with higher environmental quality values, especially forest and meadow (Table 8 and Table 9), offsets the negative impacts brought by the relatively small expansion trends in production and living space. The trade-off in this way makes the overall levels of quality of the ecological environment under these two scenarios slightly improved from those of 2020. It should be noted that if it were the case that the perception of ecological benefits and ecosystem service were different for each land-use type, the results with respect to ecological benefit might vary.

4. Discussion

This study, based on the available and relevant policies associated with land-use planning (Table 3), developed three scenarios (S1, S2, and S3). With the business-as-usual scenario as the baseline (reference scenario, RS), four possible outcomes are determined within the multi-scenario simulation for the Yubei district. Overall, the scenarios developed satisfy most policies while emphasising land use relative to different aspects of sustainable development.

The aim of this case study is to demonstrate the applicability and benefits of such a multi-scenario simulation in decision-making processes under current policies and the contemporary political agenda. The discussion thus focused on the considerations and challenges encountered during the scenario development and simulation set-up, as well as the question of how the simulated results can aid decision-making in future planning.

4.1. Developing Policy-Driven Scenarios

Scenario S1 is developed in the context of emphasizing economic development (Section 3.2.2). While the parameters used in the matrices in the simulation have been set to abide by all the policies considered in this study, it is evident that the prioritization of rapid urban expansion and infrastructure development still leads to significant encroachments on ecological and agricultural spaces. While the scenario aims to enhance socio-economic benefits through increases in construction land, it poses a certain risk to food security by reducing the production area. In addition, the relatively generous parameter setting that allows ecological space to be converted to other spaces may not be in line with the principles of ecological civilization development and the strategic goals outlined in local governance, which emphasize the need to consider the balanced growth of ecological integrity. This also overlooks the potential ecosystem service of enhanced social well-being.

The parameter setting in the green development scenario (Section 3.2.3), conversely, emphasizes ecological preservation, going beyond simply meeting the minimal requirements of contemporary environmental standards and planning guidelines. By restricting the conversion of ecological spaces into developed land, the setting of this scenario shows support for long-term ecosystem integrity within the region. In this scenario, the eco-environmental quality index exceeds the value current in 2020 and is the highest among all scenarios. This scenario demonstrates that prioritizing ecological integrity can lead to measurable improvements in ecological quality. It underscores the importance of integrating ecological objectives into land-use planning. Notably, the changes in construction space and production space in this scenario are within the acceptable level required by the related policies.

The cultivated-land protection scenario (Section 3.2.4) presents a possibility for agricultural sustainability, effectively addressing the critical need to protect cropland resources. By increasing the costs associated with converting cultivated land into other uses, this scenario reinforces the importance of agricultural viability in the face of urbanization pressures. This aligns with established policies aimed at preserving farmland and highlights the necessity of integrating agricultural land-use considerations into broader regional planning efforts.

We have demonstrated that the setup of the multiple policy-driven scenarios using adjusted parameters not only reflects the policy focuses but also meets minimal requirements of all relevant policies. This outcome shows that the decision-makers possess a degree of flexibility in implementing land-use policies, provided they aim to satisfy the basic conditions outlined in the different regulations. Yet, we have demonstrated that different policy focuses can lead to markedly different visions of city development. Collectively, by setting scenarios with varying emphases across aspects of sustainable development, this study underscores the importance of harmonizing economic, social, and ecological objectives in the Yubei District. Each scenario offers valuable guidance for future policymaking, ensuring that development strategies adhere to the principles outlined in the 14th Five-Year Plan and the long-term vision for 2035, as well as the TSP of the Yubei District. Specifically, the simulation results of the RS and S1 scenarios suggest that if development continues unchecked or prioritizes economic growth, as reflected in the ease of converting various land types into construction land, the Yubei District will struggle to meet the agenda of development according to the ECCDA framework, despite fulfilling TSP and long-term targets.

The policy-driven scenario development here demonstrates that, even while satisfying all major policy requirements, there remains space for negotiation in planning decisions. Within such room for negotiation, decision-makers need to prioritize among different aspects of sustainability, and the results of the multi-scenario simulation provide a meaningful reference to guide these choices. The favorability of each scenario depends on which sustainability dimensions are prioritized. In particular, under the current political agenda of ecological civilization, environmental sustainability may be prioritized, making the green development scenario (S2) more favorable. Alternatively, should food security become a pressing concern, the agricultural-land protection scenario (S3) may be emphasized to increase the overall resilience of the city.

Following the results of the simulation, this study has applied the eco-environmental index to evaluate the effects of land-use changes. The use of this index has highlighted that the study places particular value on maintaining environmental quality, especially the integrity of the natural environment, above other, more anthropogenic development objectives. This aligns with broader ideological shift toward ecological civilization in China, aiming to recalibrate the development trajectory away from a purely GDP-driven model, and towards a more environmentally conscious path.

4.2. Applicability of Eco-Environmental Effect Assessment

This study analyzed ecological effects (Section 3.3) from the perspective of the reciprocal transformation of land-use types. The research admits that there are many other factors affecting the ecological changes, such as the map patch size and the landscape ecological index. It remains to be studied how the ecological environment will change under the influence of these factors. Compared with other evaluation methods, the eco-environmental quality index method can estimate the eco-environmental effect based on the areas of the PLE spaces, which are comparatively straightforward to calculate. As mentioned previously, a set of quality indices specifically determined for China have been established by experts and have been applied in a few studies [51,52,53,54,55,56]. Using this index in this study thus provides a strong practical advantage. The results can conveniently be discussed based on comparisons to a previous study.

Admittedly, the applicability of this approach to the valuation of ecosystem services is contentious. In most cases, the corrections of indices depend on experts’ surveys or empirical statistical models [82]. In the cases where the index is applied, it is clear that the valuation can only be used to the level of the secondary classification of the land-use classification system (Table 2). At this level, the expert opinion has proportioned the ecosystem service primarily with respect to the degree of green cover and the level of wilderness. Such evaluations may value regulatory and supporting ecosystem services such as climate regulation, flood control, water purification, nutrient cycling, soil formation, and habitat provision higher than the provision and cultural services defined in the Millennium Ecosystem Assessment [83]. Due to the complexity and uncontrollability of ecosystems affected by the natural environment and human activities, the results of this evaluation method will be questioned as to bias and uncertainty and compared to the results of site-specific investigations [84]. However, in this study, we chose this method to evaluate the eco-environmental impacts of future land-use spatial changes hoping to consider the outcomes under different scenarios from the perspectives of decision-makers, who usually prefer using evaluation criteria that have been applied before, so that they can cross-reference the results from other cases. Additionally, the index offers a straightforward and practical way to estimate the ecological effects of land-use changes, particularly in terms of the areas of PLE spaces, aligning with current policy orientation. This makes the findings accessible to decision-makers who need to quickly assess potential ecological impacts and make choices based on current policy. Further, the ease of calculation and applicability across different scenarios provide clear advantages in guiding decisions related to land-use planning and policy development, especially when expert input or empirical data might not always be readily available. As the use of this method is versatile under different scenarios, it serves the needs of decision-makers who are considering various future possibilities, utilizing various decision options. Of course, a constant monitoring of the environmental quality to validate the reliability of the evaluation results is needed, as with many other environmental assessment methods.

4.3. Policy Recommendations Under Proposed Scenarios

Scenario-based planning transcends mere forecasting [85]. It serves as a creative articulation of potential future scenarios, as derived from an analysis of historical, contemporary, and forthcoming challenges. Consequently, this systemic approach represents a significant advancement in the concept of sustainable urban development. Scenarios enable policymakers to evaluate various strategies by investigating alternative futures, alert stakeholders to uncertainties, and facilitate the formulation of a coherent vision among all involved parties [78,85]. The land-use scenarios analyzed for the Yubei District highlight the critical need for effective policy interventions to balance socio-economic development with ecological sustainability; thereby, a resilient urban environment can be maintained during the transition. The simulations conducted across the four distinct scenarios seek to provide valuable insights into the implications of different land-use policies.

The findings of RS emphasize that unchecked urbanization driven by economic needs could lead to a decline in regional eco-environmental quality, and the reduction in ecological land does not meet the development requirements of the ECCDA (i.e., the ecological function of land cannot be reduced and changed, P1 and P2). This scenario serves as a cautionary tale, indicating that without regulatory measures, the pressures of urban expansion can severely undermine the region’s ecological balance.

S1, building upon the RS, incorporates economic development policies that reduce the probability of the transfer of construction land to other uses. In this way, some controls is placed on the expansion of built-up areas. It is worth to indicated that, this scenario still sees decreases in agricultural and ecological spaces, and notable growth in the area associated with industrial and mining production. This phenomenon underscores the fact that only targeted interventions that promote sustainable economic growth can safeguard vital ecological resources, though they might not necessarily restore the already-compromised environmental quality. The outcome of this scenario does not align with the ECCDA’s requirement of not reducing the ecological land; however, it is consistent with other development policies and plans associated with the Yubei District. To address the impact of construction land expansion on ecological space, a combination of mandatory ecological compensation and urban green infrastructure interventions can be implemented to meet the ECCDA’s “no net loss of ecological function” requirement. Ecological compensation mechanisms effectively offset the ecological losses caused by development, such as through payments supporting ecological restoration or conservation projects [86]. At the same time, green infrastructure (such as urban green spaces, green roofs, and sustainable drainage systems) can enhance ecological connectivity in cities, providing multiple ecosystem services, such as biodiversity protection, water purification, and air quality improvement [87]. By incorporating these measures into urban planning, the negative impacts of construction activities on the natural environment can be minimized, ensuring the continuity of ecological functions [88].

In contrast, the S2 scenario, aligned with the principles of green development, indicates that achieving a balanced dynamic between production and ecological spaces is possible, based on the setting of conversion probability. The slight increases in forest, meadow, and water ecological spaces illustrates the effectiveness of implementing policies that restrict the conversion of ecological areas into urban developments. The environmental quality index, based on the simulation outcome of this scenario, is thus increased. This scenario advocates for a comprehensive approach that emphasizes the integration of ecological considerations into planning frameworks. To achieve this desirable condition, policies should foster the establishment of protected areas and promote sustainable land-use practices that accommodate both economic growth and environmental preservation. For example, AI-driven simulations could be utilized to predict the impact of urban planning on both economic growth and ecological protection. By adjusting land-use policies and promoting sustainable development strategies, such as green industry growth and efficient resource use, they demonstrate how cities can achieve economic growth without compromising ecological health [89].

S3 further reinforces the importance of agricultural-land protection, resulting in a minimal decrease in agricultural production space, while maintaining ecological integrity. This scenario mitigates the pressures of urbanization on farmland resources. China has always prioritized food security-related land-use policies such as farmland protection, and the prevention of non-agricultural use of agricultural land. Although there has been controversy over the correlation between intact farmland protection and food security in rapidly urbanized areas like the Yubei District [90], this scenario can still provide a reasonable evaluation for researchers and policymakers to use to address potential evolutions in the trends associated with land use when farmland protection policies are prioritized.

In evaluating the eco-environment quality index across the scenarios, it becomes evident that the results of S2 and S3 exhibit the highest levels of ecological health, reinforcing the argument for prioritizing green policies and agricultural protection measures. Meanwhile, these two scenarios also comply with all land-use policies and ECCDA-related regulations in the Yubei District. These scenarios indicate that limiting the expansion of production and living spaces results in improved ecological conditions. Therefore, comprehensive policy frameworks must be designed to facilitate the transition toward more effective use of existing living and production areas so that the sustainable land-use practices can be achieved.

Overall, under the context of ecological civilization, scenarios S2 and S3 serve as viable reference frameworks for the future development of the Yubei District. Given that the Yubei District is still in the stages of urbanization and rapid economic growth, there is a substantial demand for production and living spaces [91,92]. Therefore, it is imperative for the region to enhance land-use efficiency with respect to its industrial production and living functions, establish robust ecological boundaries, and ensure the protection of essential farmland from encroachment. The objective should be to create a healthy urban environment characterized by ecological integrity, livability, and high production efficiency. In this way, the development of the district can be self-sustaining and resilient to external changes.

4.4. Research Limitations and Future Work

The primary limitation of this study is the subjective nature of the scenario-setting methods. While different scenarios can indicate the likelihood of associated land-use changes, they do not necessarily reflect the actual future land-use patterns. To overcome this limitation, this study selects additional driving factors and employs multiple scenario-setting methods to conduct multifactorial and multi-scenario land-use change simulations. This approach aims to yield a more holistic insight into potential land-use alterations, which could more precisely answer the “what-if” questions under varying conditions, thereby assisting policymakers in the development of more effective land-use strategies. Comparing the actual land use and the simulated results in Yubei (2020), the OA and Kappa, the parameters used for evaluating the accuracy of the modelling, as well as the visual examination of projected results, indicated that the modelling efforts can produce reasonable predictions.

Another limitation of this study is the exclusive use of the classic Markov–FLUS model. While it is acknowledged that the Markov–FLUS model is well-suited for the study area, its assumptions (e.g., transition probabilities) may limit accuracy in areas with highly dynamic or unpredictable land-use patterns. The latest research shows that some models, such as machine learning-enhanced simulation, produce more accurate and precise results in their capturing of the complexity of land-use change in a given area, compared to the traditional Markov–FLUS model [93,94]. Therefore, using other models may have an impact on the simulation results. By integrating new land-use simulation models, future research can provide more comprehensive and accurate insights into the dynamics of land-use change.

Overall, this research makes contributions to the domains of policy-oriented scenario development and land-use change modelling. Nevertheless, recognizing and mitigating its limitations is essential to enhancing the validity and dependability of the results. By integrating sophisticated models and utilizing diverse scenario-setting approaches, subsequent investigations can offer more thorough and precise insights into the dynamics of land-use change, thereby aiding in the promotion of sustainable land-use planning and management practices.

4.5. Research Implications

This case study in the Yubei District offers practical insight on how to make land-use decisions balancing urban development, ecological preservation, and agricultural protection within policy boundaries. The use of policy-based multi-scenario simulations in the Yubei District demonstrates the value of exploring different land-use strategies while adhering to overarching policies. For countries facing rapid urbanization, this approach can help policymakers identify sustainable pathways that integrate ecological considerations into development planning.

The study highlights that it is possible to prioritize environmental protection without hindering economic growth, which is a critical viewpoint for developing countries and regions. Additionally, the flexibility within the policy frameworks allows for specific strategies that address local challenges while meeting long-term sustainability goals. Overall, the Yubei case provides an example of the means of proposing adaptable, policy-driven land-use strategies that balance socio-economic and ecological objectives.

5. Conclusions

This study examined the shifts in production–living–ecological (PLE) space in the Yubei District from 2005 to 2020, and projected land-use changes through 2035, under one reference and three scenarios. The findings highlight significant transformations marked by substantial expansions of living space, declines in production space, and relative stability of ecological areas. The conversion of agricultural land, with its higher eco-environmental quality index compared to urban and industrial territories, into urban and industrial zones underscores the tension between urban development and ecological sustainability. The four scenarios analyzed—the reference scenario (RS), economic development scenario (S1), green development scenario (S2), and agricultural land protection scenario (S3)—reveal a spectrum of implications stemming from varied land-use policies. The RS scenario, characterized by unrestricted transformations, demonstrates detrimental impacts on shrinking agricultural land and ecological spaces, while S1 reinforces urban expansion with ecological costs, necessitating strategic interventions to rebalance economic and environmental goals under the agenda of ecological civilization.

In contrast, S2 and S3 emerge as viable pathways aligned with ecological civilization-based principles, showcasing enhanced ecological quality and the necessity of preserving natural habitats amidst socio-economic imperatives. S2 advocates for a balanced growth model that maintains ecological integrity, whereas S3 emphasizes the protection of agricultural land to ensure food security. These scenarios illuminate the need for robust regulatory frameworks and the establishment of ecological boundaries to protect vital farmland and ecosystems. The eco-environmental quality index assessments affirm that S2 and S3 achieve increased scores for ecological health, illustrating the effectiveness of land-use decision-making based on proposed scenarios.

This case study demonstrates that policy-driven multi-scenario simulation can offer meaningful guidance for planning decisions. However, we recognize that simulation models often assume that historical trends will continue. Experience suggests that as a country transitions from a developing to a more mature economy, growth rates tend to slow, household sizes decline, and birth rates drop. In light of these dynamics, we argue that the baseline trajectories in land-use simulations may need regular updating. To better support decision-making under these changing circumstances, land-use modeling based on policy-driven scenarios should integrate additional tools—such as demographic and macroeconomic projections—to dynamically assess the evolving requirements for living, production, and ecological spaces.

Collectively, the findings of this study advocate for integrated land-use strategies that harmonize ecological, economic, and social objectives. As the Yubei District continues to urbanize, adhering to the principles of balanced, resilient and ecologically concise development will be paramount in safeguarding its ecological assets while meeting the demands of a dynamic population.