Sparing or Sharing? Differential Management of Cultivated Land Based on the “Landscape Differentiation–Function Matching” Analytical Framework

Abstract

The sole function of cultivated land of agricultural production is insufficient to meet the diverse demands of modern agriculture. To address land-use conflicts and achieve the United Nations Sustainable Development Goals (SDGs) of zero hunger and reduced carbon emissions by 2030, this study introduces the theory of land sparing and sharing, uses landscape indices to identify spatially fragmented areas, employs a four-quadrant model to assess the matching status of functional supply and demand, and applies correlation analysis to determine the trade-off/synergy relationships between functions. The results indicate the following: (1) Zhengzhou’s farmland landscape exhibits characteristics of low density, low continuity, and high aggregation, with separation zones and sharing zones accounting for 77% and 23% of the total farmland area, respectively. (2) The multifunctional supply (high in the northeast, low in the southwest) and demand (high in the west, low in the east) of farmland show significant mismatches, with PF and EF exhibiting the most pronounced supply–demand mismatches. The “LS-LD and HS-LD” types of farmland account for the largest proportions, at 39% and 35%, respectively. (3) The study area is divided into four primary types: “PCZ, RLZ, BDZ, and MAZ” to optimize supply–demand relationships and utilization patterns. This study enriches the application of land sparing and sharing in related fields, providing important references for policymakers in optimizing land-use allocation and balancing food and ecological security.

1. Introduction

Cultivated land constitutes a vital resource for global material production, supporting approximately 85% of humanity’s food supply. According to a 2024 FAO report, global progress toward hunger eradication has stagnated or even regressed, with an estimated 733 million people currently experiencing chronic undernourishment [1,2]. Furthermore, food insecurity risks exacerbating geopolitical tensions, particularly in contexts of escalating natural disasters, regional conflicts, and export restrictions imposed by key agricultural nations [3,4,5]. These dynamics underscore the urgent imperative to prioritize resilient and equitable food systems as a cornerstone of international stability.

Land sparing and land sharing, conceptualized by Green et al. [6], has become a key strategy for reconciling interests in global environmental challenges. Recent studies highlight their divergent ecological and socio-economic implications across ecosystems. In tropical Colombia and the Amazon [7,8], land sparing retains significantly higher agricultural sustainability effects than land sharing, with 34.80% and 25.20% higher levels of high and low yields, respectively, and 22.50% and 18.93% higher functional richness, regardless of the level of landscape wildlife friendliness, whereas composite strategies prove more effective in Southeast Asian urban parks and European cities [9,10], land-sharing urban areas are significantly associated with a higher taxonomic and functional diversity of birds during winter. Canadian forestry research reveals trade-offs between old-growth forest retention and infrastructure development, involving more than 50 percent of targeted biomass harvesting, more than doubling road densities and more than tripling construction and maintenance costs when uneven-aged management is used [11], while in the case of fire in forests and grazing in semi-natural grasslands, land sharing complements sparing for species vulnerable to natural disturbances [12]. Economically, UK-cultivated land analyses suggest that sparing optimizes cost-efficiency in meeting biodiversity and carbon targets, since sparing delivers our biodiversity and carbon emission target outcomes at 79% of the food production cost and 48% of the taxpayer cost of sharing [13], aligning with the international vision of the Sustainable Development Goals (SDGs) [14]. Contrastingly, Brazilian studies emphasize human well-being dimensions in strategy design, as little is known about the comparative effects of different interventions based on land sharing or land separation on local definitions of human well-being [15]. China’s Maoming region demonstrates superior ecosystem service integration under land sharing, and while integrated ecosystem services were found to be highest in the land-sharing model, food provisioning services declined to varying degrees when the share of arable land area exceeded 84.84% [16], and critics note framework limitations, including an overreliance on yield–density assumptions [17]. Despite this, hybrid approaches show promise in balancing environmental and biodiversity outcomes [18,19]. Mechanistically, sparing prioritizes high-density, low-heterogeneity land use, while sharing favors low-density, high-heterogeneity patterns [20]. Thus, the scientific delineation of the spatial allocation of the two strategies is crucial for the synergistic advancement of production and ecology.

Multifunctional agriculture, initially advanced by the European Union, has gained global traction as land’s productive, ecological, esthetic, and social roles intensify [21,22,23]. Driven by evolving societal needs, many nations now prioritize multifunctional cultivated land management. Case studies illustrate this shift: research in the Wuhan metropolitan area in China aims to mitigate the trade-off between food security (SDG 2) and land degradation (SDG 15) [24]. Economic and demographic urbanization is a major factor in the mismatch between multifunctional supply and demand for arable land in rapidly urbanizing areas [25]. Specifically, the transformation of agricultural land use for production and subsistence functions stimulates local and neighboring regional food production [26,27]. Although integrated agriculture has economic advantages (ωF = 0.377), ecological agriculture contributes more to meeting social needs (ωF = 0.623) [28]. The increase or decrease in arable land in different agro-climatic zones has led to prominent cross-regional farmland problems [29]. Especially in the Yellow River Basin, the multifunctional conflict of ecological protection–agricultural production–urban construction accounted for 66.33% [30], and non-agricultural and non-food diversion needs to be strictly controlled [31]. In northern China, on the other hand, yield enhancement has benefited from elements such as planting intensity and irrigated area [32]. Cropland functional trade-offs will intensify in the future [33], and there is an urgent need for the government to develop production and landscapes while reducing the loss of cropland value due to pollution [34]. It is worth noting that landowners prefer to retain rather than convert farmland [35], but still face several challenges: suitable area planning, maintenance management, production optimization, and technical support [36]. Therefore, it is important to scientifically integrate future demand for goods and services [37], aligning with China’s 2025 goal of fully harnessing agricultural multifunctionality [38]. Furthermore, innovations in agricultural productivity amplify cultivated land value [39], positioning multifunctional supply–demand alignment as critical for resolving land use conflicts and advancing rational spatial planning [37].

Existing studies lack empirical research from the perspective of combining supply and demand, ignoring the two-sided and systematic character of the function of cultivated land. The land sparing and sharing framework, which is gradually being used to harmonize ecological protection of the environment with the multiple needs of human beings. Optimization and implementation of land use decisions should also consider landscape heterogeneity and stakeholders.

Therefore, the aim of this paper is to apply the land sparing and sharing strategy to the management of cultivated land conservation. The key research question is to identify the fractured space of cultivated landscapes and the interrelationship of functions. Zhengzhou City has a prominent human–land conflict, with its six main agricultural areas as the study area. Cultivated land resource use is affected by a combination of natural–environmental and social–economic systems. Cultivated land resource use is affected by a combination of natural–environmental and social–economic systems (Figure 1). The first is the use of landscape differentiation to identify fractured spaces, with the result reflecting a “single or multiple”. The second is the use of multi-functional matching to identify equilibrium conditions, with the result reflecting a “surplus or deficit”. The third is the use of correlation analyses to identify sub-functional relevance, with the result reflecting “trade-offs or synergies”. Finally, zoning optimization and restructuring were carried out and differentiated control measures were proposed.

2. Materials and Methods

2.1. Study Area

Zhengzhou City is located in the middle and lower reaches of the Yellow River in China (112°42′114°14′ E, 34°16′34°58′ N), and is the capital city of Henan Province, which has a north-temperate continental monsoon climate, with a topography that is high in the west and low in the east (Figure 2). With an average annual precipitation of 623.3 mm and an average annual temperature of 14.2 °C [40], the soil is fertile and suitable for a variety of crops. Dry land accounts for 33 percent of Zhengzhou’s cultivated land, mainly in the mountainous hills of the southwestern part of the city; water-consumed land accounts for 68 percent, mainly in the eastern plains. As an important provincial capital city in the middle reaches of the Yellow River, it faces conflicts between ecological protection, food security, and high-quality development. To better meet the multiple demands for land resources under urban renewal, rational planning and management of cultivated land are the keys to balancing urban development and regional food security.

2.2. Data Sources

Land use data from China Land Cover Dataset (CLCD) (https://www.earth-system-science-data.net); digital elevation model (DEM) data with a spatial resolution of 30 m from the ASTER GDEMV3 dataset (https://www.gscloud.cn, accessed on 5 March 2024); population data from China Grid Population Dataset (CGPD) (https://www.earth-system-science-data.net, accessed on 20 March 2024); soil property data from the Earth Resources Data Cloud Platform (ERDP) (www.gis5g.com, accessed on 15 May 2024); normalized difference vegetation index (NDVI) data from the National Tibetan Plateau Science Data Centre (https://data.tpdc.ac.cn, accessed on 23 June 2024); net primary productivity (NPP) data from NASA Surface Processes Data Centre 500 m MOD17 Dataset (https://www.earthdata.nasa.gov, accessed on 13 July 2024); national rural tourism key villages data from the official website of China’s Ministry of Culture and Tourism (https://www.mct.gov.cn/, accessed on 20 August 2024), and spatial points were identified through a geographic counter-checking coordinate tool. Cultivated land attribute data from the 2012 Zhengzhou City Agricultural Land Force Evaluation results. Data in the socio-economic and agriculture-related categories are derived from the Zhengzhou City Statistical Yearbook and statistical bulletins on national economic and social development in each region. All geographic images were uniformly map-projected and coordinate-corrected using the CGCS2000 coordinate system, and resampled to a grid with a spatial resolution of 30 m × 30 m.

2.3. Research Methodology

The use of cultivated land resources is influenced by both the natural environment at the supply level and the socio-economic system at the demand level. This study follows a logical chain of “data integration, model selection, status characterization, and management measures”. First, Fragstats was used to analyze the landscape fragmentation characteristics of farmland, and the spatial distribution of separated and shared areas was defined according to certain classification criteria. Second, a multifunctional supply–demand evaluation index system for farmland is constructed using multi-source data such as soil and NDVI. Through supply–demand matching analysis, farmland is classified into four types. Subsequently, correlation analysis is conducted to determine the degree of association between sub-functions, reflecting the level of significance. Finally, zone types are delineated, and differentiated measures and optimized control pathways are proposed to support the achievement of the United Nations Sustainable Development Goals (SDGs) (Figure 3).

2.3.1. Landscape Character Recognition

The landscape index can be used to identify regional landscape fragmentation, connectivity and agglomeration and other related information [41], the selection of landscape indices follows the principles of relevance, independence, and academic consensus. PD is defined as the number of cropland patches per unit area, which can directly characterize the degree of cropland fragmentation. AI measures the degree of spatial distribution of cropland patches in a concentrated manner, and CONNECT assesses the potential of material and energy flow through the spatial adjacency between patches. The three are widely used in the study of cropland and wetland. The redundancy of indicators was eliminated through correlation analysis to avoid the problem of multiple covariance. (

Table 1. Calculation and meaning of landscape indicators.

Type of Indicator	Calculation Method	Implications of the Indicator
PD	$PD=\left[n_i/A\right](1000)(100)$	Reflecting fragmentation between patches, the larger the PD, the more fragmented the landscape tends to be, leading to diversified use.
CONNECT	$CONNECT=\left[{\displaystyle \frac{{\displaystyle {\sum }_{j=k}^n{c}_{ijk}}}{\frac{n_i(n_i-1)}{2}}}\right]\times 100$	Reflecting structural continuity, the larger the CONNECT, the more continuous the landscape tends to be, leading to single use.
AI	$AI=\left[{\displaystyle \frac{g_{ii}}{\mathit{max}\to g_{ii}}}\right](100)$	Reflecting the clustering between patches, the smaller the AI, the more dispersed the landscape tends to be, leading to diversified use.

).

2.3.2. Evaluation System Construction and Matching Supply and Demand

(1) Multifunctional Evaluation

The construction of the evaluation indicator system follows some principles of systematicity, representativeness, and operability. This paper combines relevant studies with the actual natural conditions and economic development of Zhengzhou City to select suitable evaluation indicators from both supply and demand (

Table 2. System of multifunctional supply and demand indicators for cultivated land.

Type	Level	Indicator	Weights	Method of Calculation	Reference
Production function	Supply	Crop yield (Cy)	0.402 (+)	Cy = (production potential score of grid i/total value of production potential of region j) * total actual production of cultivated land in region j. Grid i makes up region j.	[42,43]
		Soil quality (Si)	0.598 (+)	Comprehensive assessment combining attribute characteristics of grid i such as “geomorphology, texture, slope, organic matter, effective phosphorus and potassium, irrigation, and drainage capacity”.	[43,44]
	Demand	Economic driver index (Edi)	1 (+)	Gross domestic product (GDP) at 1 km resolution was used as a proxy for crop consumption data. It is assumed that a higher GDP is associated with higher food quality requirements, indirectly reflecting demand.	[42,45,46]
Ecological function	Supply	Soil conservation (Ac)	0.551 (+)	Ac = R * K * LS * (1 – C * Ps), the Revised Universal Soil Loss Equation (RUSLE) was used in this study for computational assessment.	[25,28,47]
		Soil sequestration (Cs)	0.449 (+)	Cs = 1.63 * NPPi + SOCDi, NPPi and SOCDi are the dry matter accumulation capacity and soil organic carbon density of grid i, respectively.	[43,48,49,50]
	Demand	Neighborhood Ecological Index (Nei)	1 (−)	Measured by the percentage of ecological land (woodland, shrubs, grassland, water) within 1 km of cultivated land. It is assumed that the lower the percentage, the greater the ecological demand for farmland.	[51]
Landscape cultural function	Supply	Leisure and entertainment (Le)	0.396 (+)	Google Maps extracts POIs such as “picking gardens, farms, leisure farms” and takes the 5 km range of farmland as the core area of agritourism.	[25,37]
		Green environment (Ge)	0.604 (+)	Ge = (NDVI-NDVIs)/(NDVIv-NDVIs), NDVI, NDVIv, and NDVIs are vegetation indices, vegetation indices for pure vegetation, and vegetation indices for pure soil, respectively.	[33,52]
	Demand	Benefit level of the population (Rbl)	1 (+)	Considering the impact of proximity on sightseeing trips, the distribution of beneficiaries is characterized. It is assumed that the denser the PD and the closer it is to the key villages of rural tourism, the higher the level of demand from residents.	[43,53]
Social security function	Supply	Farming income (Fi)	0.620 (+)	An indicator of rural residents “net household business income was selected from the Zhengzhou City Statistical Yearbook to reflect the role of cultivated land in supporting farming economy.	[27]
		Employment security (Es)	0.380 (+)	Es = population working in agriculture/total area of cultivated land in the region, the result of this calculation shows the role of cultivated land in carrying agricultural laborers.	[28,54]
	Demand	Level of food distribution (Frl)	1 (+)	Frl = Foodi/(400 * POPi), Foodi and POPi are the crop production of grid i and the population of grid i, respectively; 400 is a safe value for per capita food consumption.	[26,33,53]

Note: (+) indicates that the indicator is positive; (−) indicates that the indicator is negative; * indicates that the variables are multiplied.

). The indicators were standardized by the min–max method, the entropy method was used to assign weights to the indicators, and the comprehensive evaluation method was used to calculate the scores for each function.

$$x_{\mathit{ij}}^{*}={\displaystyle \frac{x_{ij}-x_{\mathit{min}}}{x_{\mathit{max}}-x_{\mathit{min}}}}/{\displaystyle \frac{x_{\mathit{max}}-x_{ij}}{x_{\mathit{max}}-x_{\mathit{min}}}}$$

(1)

$$W_j={\displaystyle \frac{1-e_j}{{\displaystyle \sum _{j=1}^{m}1-e_j}}}$$

(2)

$$F_k={\displaystyle \sum _{i=1}^n{W}_j}\times x_{\mathit{ij}}^{*}$$

(3)

where $x_{\mathit{ij}}$ is the value of the $j$-th indicator for the $i$-th cell; $x_{\mathit{min}}$ and $x_{\mathit{max}}$ represent the minimum and maximum values in the data respectively; $x_{\mathit{ij}}^{*}$ is the standardised value of the $j$-th indicator for the $i$-th cell; $e_j$ is the entropy value of the $j$-th indicator; $W_j$ is the weight of the $j$-th indicator; $F_k$ is the composite assessment value.

(2) Supply and Demand Matching

Constructing a four-quadrant model to measure the state of matching supply and demand for cultivated land (surplus, equilibrium, or shortage). Using the composite supply index and the composite demand index as the horizontal and vertical axes of the coordinate system, respectively, the mean values of the multifunctional supply and demand composite indices for cultivated land were used as the midpoints of the horizontal and vertical axes, respectively, to classify them into four types of matches (Figure 4).

2.3.3. Analysis of Trade-Offs/Synergies Between Multifunctionality

Coupled Coordination Model Building

The coupled coordination degree model is used to describe whether the interactions, integration efficiencies, or combined benefits of multiple systems or elements are coherent. The indicator parameters are embedded into the expanded model and classified into three types, high, medium, and low, with the help of the natural breakpoint method. The modified formula is:

$$C=\sqrt[4]{F(pro)F(eco)F(lan)F(\mathit{soc})/{\left(\left(F(pro)+F(eco)+F(lan)+F(\mathit{soc})\right)/4\right)}^4}$$

(4)

$$T=\alpha F(\mathit{pro})+\beta F(\mathit{eco})+\gamma F(lan)+\delta F(soc)$$

(5)

$$D=\sqrt{C\times T}$$

(6)

where $F\left(pro\right)$, $F\left(eco\right)$, $F\left(lan\right)$ and $F\left(sco\right)$ represent the values of the productive, ecological, landscape cultural and social security functions of cultivated land; $\alpha$, $\beta$, $\gamma$, and $\delta$ represent the weights of four functions; $C$, $T$ and $D$ represent degrees of coupling, coordination and coupling, and coordination.

Spearman correlation analysis

Spearman’s analysis is used to measure the dependence between two variables, the direction of the trend of change and the strength of the correlation. In this paper, the method is used to explore the interrelationships between the sub-functions of cultivated land, where the positive correlation coefficient and the negative correlation coefficient denote the synergistic and trade-off relationships between the sub-functions, respectively. It is known that W and V are sequences of two sets of variables, respectively, and the formulas are as follows:

$${\rho }_s={\displaystyle \sum _{i=1}^N\left(W_i-\overline{W}\right)\left(V_i-\overline{V}\right)}/\sqrt{{\displaystyle \sum _{i=1}^N{\left(W_i-\overline{W}\right)}^2{\displaystyle \sum _{i=1}^N{\left(V_i-\overline{V}\right)}^2}}}$$

(7)

where $N$ is the total number of observations; $W_i$ and $V_i$ are the ranks assigned based on the value of observation $i$, respectively. $\overline{W}$ and $\overline{V}$ are the mean ranks of the variables $W$ and $V$, respectively. ${\rho }_s$ is the correlation coefficient.

3. Results

3.1. Landscape Pattern Analysis

According to the analysis of the calculation of the cultivated landscape index in Figure 5a–c. For patch density, the southwestern region (Xinmi, Gongyi, Dengfeng), with great topographic relief, has a high degree of fragmentation of cultivated land, and agricultural production exhibits diversified utilization patterns. The eastern Yellow Huaihai Plain (Zhongmou and Xinzheng) has a relatively gentle topography and better integrity and regularity of cultivated land, making it suitable for large-scale production. For the connectivity index, the high-value areas are mainly distributed in the southeastern region, which is mainly caused by the distance between the cropland patches, and the greater the connectivity (Zhongmou and Xinzheng), the stronger the effect of mutual constraints and facilitation between the cropland patches. For the aggregation index, the high-value areas are mainly located in the northern region along the Yellow River and the eastern plains, where the terrain is gentle, water resources are abundant, and the distribution of land parcels is relatively aggregated, which facilitates the efficient integration and utilization of resources. Based on the above analysis, combined with certain partitioning criteria, categorization statistics are shown in

Table 3. Rules for classifying landscape types of cultivated land.

Partition Type	Number of Patches	Delineation Criteria
Relative Sparing Zone	571	AI > 75 and CONNECT > 40 and PD < 0.015
Sparing Marginal Zone	619	AI > 75 and CONNECT ≤ 40 or AI > 75 & CONNECT > 40 and PD ≥ 0.015
Relative Sharing Zone	349	AI ≤ 75 and CONNECT > 40 or AI ≤ 75 & CONNECT ≤ 40 and PD < 0.015
Sharing Marginal Zone	535	AI ≤ 75 and CONNECT ≤ 40 and PD ≥ 0.015

, and the partitioning results were visualized in Figure 5d.

3.2. Multifunctional Evaluation Analysis

The comprehensive supply level of cultivated land, in general, shows a spatial pattern of high in the northeast and low in the southwest (Figure 6). The areas with high PS values are located in Zhongmou in the east and Xingyang in the north, while Dengfeng, Gongyi, and Xinmi in the southwest, which are in the remnant section of the mountainous terrain, have lower PS values for cultivated land. ES was generally at a high level, roughly higher in the hilly areas than in the plains and higher in the areas along the Yellow River than in the non-Yellow River areas, which indicates that the soil retention and carbon sequestration capacity showed a geographically differentiated status. LS shows a ‘gradient’ distribution from the center to the periphery, with high values located in areas along the Yellow River, such as Xingyang, as well as in Zhongmou and Xinzheng, which are located in the Yellow–Huaikai Alluvial Fan Plain. The SS high-value zones are then distributed on both sides of the municipality, with low-value zones in the center.

The combined level of demand for cultivated land generally shows a spatial pattern of high in the west and low in the east. The high-value areas of the PD are mainly located in the centers of the counties, while the low-value areas are the peripheral zones deviating from the center of the towns. The spatial distribution characteristics of ED and ES are similar, and PD often shows high values due to the problem of small and scattered cropland areas in hilly areas, where connectivity and agglomeration effects are weak. LD is generally at a high level, with high-value areas located in zones close to urban areas where there is a strong demand for the esthetic value of the landscape of cultivated land. SD shows a spatially decreasing trend from west to east, which is closely related to regional agricultural production and population distribution.

3.3. Supply and Demand Matching Analysis

A radar chart showing the differences in the level of supply and demand for cultivated land functions by county according to the statistics (Figure 7). From the horizontal comparison, the high-value area of ED is in Dengfeng, the high-value area of LD is in Gongyi, the high-value area of ES, LS, and PD is in Xinzheng, and the high-value area of PS, SS, and SD is in Zhongmou. In terms of vertical comparison, the PS-PD combination has a large difference in Zhongmou, Gongyi, and Xinmi, the ES-ED combination has a large difference in Xinzheng, Xinmi, and Xingyang, the LS-LD combination has a large difference in Zhongmou and Xinmi, and the SS-SD combination has a large difference in Xinzheng. In relative terms, the mismatch between supply and demand for production and ecological functions is most pronounced, while the area with the most balanced supply and demand for each function is Xingyang.

In terms of spatial distribution, the supply–demand matching (SDM) between the functions of cultivated land shows significant heterogeneous characteristics in different regions (Figure 8). “LS-HD” cultivated land (20 percent of the total) is mainly distributed in areas where map spots are heavily fragmented and the clustering effect is not obvious. “LS-LD” cultivated land (39 percent of the total) is relatively balanced in terms of the functions it performs. The “HS-HD” cultivated land (6 percent of the total) is mainly distributed along the yellow belt in the northern part of Zhongmou County, which may be influenced by the richness of the land types and the diversity of the terrain. The “HS-LD” cultivated land (35 percent of the total) has a strong agricultural base, with synergistic development of various functions, and has great potential for development.

3.4. Sub-Functional Correlation Analysis

Due to the restrictive nature of some indicator scales, six supply indicators of (Cy, Si, Ac, Cs, Le, Ge) were selected, combined two by two, and Spearman rank correlation analyses were carried out to plot the correlation coefficient matrices (Figure 9). In terms of the types of associations, of the 15 sets of associations between the six sub-functions, 8 were positive and 7 were negative. Specifically, the Cy-Si and Si-Ac combinations show a strong negative correlation, and the price of higher crop yields is loss of organic matter and structural degradation, which is caused by high-intensity agricultural practices and is closely linked to the excessive application of chemical fertilizers and monoculture. In addition, the intensive use of arable land driven by urbanization has reduced surface vegetation cover, weakening the soil’s resistance to erosion and water-holding capacity. The Cy-Ac combination shows a strong positive correlation, which is consistent with the above reasoning. EF and LCF show a synergistic relationship, which mainly benefits from the implementation of the eco-agriculture policy in Zhengzhou City, and the proactive integration of ecological functions in agritourism planning.

Four demand indicators of (Edi, Nei, Rbl, and Frl) were selected and combined two by two to plot the circular correlation coefficient. For the positive correlation, the Nei-Frl combination > Edi-Rbl combination, indicating a weak positive relationship between land use around cultivated land and the level of food security, the level of economic development and status of rural tourism, and a more efficient use of resources. For the negative correlations, the Rbl-Frl combination > Nei-Rbl combination > Edi-Nei combination > Edi-Frl combination, with strong trade-offs between most of the functions and insignificant spillover effects.

3.5. Functional Partitions and Management Paths

The results of the heterogeneity of cultivated landscapes and the matching of supply and demand above are superimposed and analyzed, and divided into four functional zones: ‘priority protection zone, improvement and enhancement zone, equilibrium development zone, and moderate adjustment zone’. The rules for the division of the various functional zones are described as follows (

Table 4. Cultivated land types of partition and status description.

Types of Partition		Combining Units
		Landscape Differences (X) + Matching Supply and Demand (Y)
PCZ	Nuclear—PCZ	(X1,Y3), (X1,Y4)
	Buffered—PCZ	(X2,Y3), (X2,Y4)
RUZ	Interior—RUZ	(X1,Y1), (X1,Y2)
	Exterior—RUZ	(X2,Y1), (X2,Y2)
BDZ	Profound—BDZ	(X3,Y3), (X3,Y4)
	Extensive—BDZ	(X4,Y3), (X4,Y4)
MAZ	Structural—MAZ	(X3,Y1), (X3,Y2)
	Integrated—MAZ	(X4,Y1), (X4,Y2)

):

According to the above-defined rules, the natural breakpoint method is used to visualize the zoning results (Figure 10), and the multi-objective scenario is established to set up the improved scenarios according to the characteristics of the functional areas (Figure 11). The description of the specific zoning scenarios and control paths is as follows:

(1) Priority Conservation Zone (PCZ)

This area is a high-quality arable land suitable for high-standard farmland, with stability and independence, consisting of 679 patches, accounting for 48.47% of the area, and spatially concentrated in most of the cultivation areas in Zhongmou and Xingyang. The results of the landscape pattern analysis show that the density of patches in this region is small, with strong agglomeration and high connectivity, which is an inherent advantage of intensive farming and large-scale operation, and can demonstrate stronger stability when coping with extreme weather [55]. The main limiting factors in the region are Ac and Cs, which are below average, and are related to high-intensity farming driven by urbanization and low surface vegetation cover. Therefore, it is necessary to promote conservation tillage techniques such as no-till straw mulching and green manure return to the field, to build vegetation buffer zones around farmland to control soil erosion and non-point source pollution [56], and to strengthen coordination with neighboring areas for strict protection through the demarcation of red lines.

(2) Rectification and Lifting Zone (RLZ)

The region is a type of arable land that needs a little improvement to achieve efficient output, and is relatively well endowed with good soil conditions and abundant water resources. The number of arable land patches is 433, accounting for 27.43% of the area, and it is mainly distributed in most areas of Xinzheng and Dengfeng in terms of spatial pattern. The degree of patch density, connectivity, and agglomeration is relatively weaker than that of the PCZ region, but the natural endowment conditions are still superior. Constrained by the double obstacles of soil quality Si and food rationing level Frl, the ‘LS-HD’ mismatch situation occurs in some areas, which requires some measures to cultivate fertilizer and nourish the land, as well as focusing on the construction of farmland water conservancy and other infrastructures [57]. In addition to considering the improvement of its own conditions, attention should also be paid to the spillover benefits of synergistic optimization with the surrounding environment. The construction of a number of model farmlands that can be watered in droughts, drained in floods, plowed, and increased in productivity is necessary.

(3) Balanced Development Zone (BDZ)

A mixed-use model is suitable for the cultivated land in this region, including 335 cultivated land patches, accounting for 9.42% of the area. Spatially, it is mainly distributed in most areas of Gongyi and Xingyang, and as an overlapping area of the “ecological barrier area” and “food production functional area” of Zhengzhou National Central City, it is suitable for the “agriculture + culture and tourism” mode and the “integrated model exploration” [58]. In contrast to the characteristics of the PCZ arable land, this area has low aggregation, weak connectivity, and low patch density. Although the supply and demand of arable land functions are relatively balanced, it is limited by the factors of recreation and green environment, and needs to be adapted to the local conditions, focusing on the development of specialty agriculture with the goal of integrating production factors and improving natural habitats.

(4) Moderate Adjustment Zone (MAZ)

The arable land in the area is not suitable for food production as arable land, and needs to be replaced by land classes to increase green space. The number of arable land plots is 502, with an area share of 14.68 percent. Spatially, they are scattered in parts of Xinmi, Dengfeng, and Gongyi, the region generally has high terrain, poor geotechnical conditions, and poor farming conditions, and EF is stronger than PF from the results. Therefore, it is necessary to change the land use mode through forest plowing or forest and grassland replacement to respond to the policy requirements of forest and fruit going up to the mountains and arable land going down to the mountains [59], and to achieve high-quality management and utilization of the resources.

4. Discussion

4.1. Strengths of the Research Framework

First, this paper proposes an analytical framework based on “landscape differentiation–functional matching”, which differs from previous studies in that it focuses singularly on the supply and demand of functions for farmland use zoning [42], or the influence of external factors on the landscape pattern of arable land [60]. This study quantitatively characterizes the state of land separation or sharing through landscape features (Figure 5) and identifies potential and obstacle zones of functions by combining the matching characteristics of the supply and demand of functions, which helps an optimization scheme based on the differentiated control of nature and society to be formulated (Figure 6), and to open up a new way to achieve SDGs. This facilitates differentiated control of the optimization scheme, which opens up a new pathway for the realization of the SDGs. Secondly, correlation analysis is the most widely used method for assessing multifunctional relationships [43], and in this paper, correlation analyses between sub-functions reveal significant synergistic relationships between Cy-Ac, Frl-Nei, and significant trade-offs between Cy-Si, Si-Ac, and Frl-Rbl (Figure 9). In addition, a coupled coordination relationship model can be constructed to further explore the interaction of multi-functionality of cropland in different regions [61].

4.2. Multifunctional Interrelationships and Alignment of Key Elements

The spatial distribution patterns of cultivated landscapes and functions derived in this paper are similar to the results of other studies of land in Zhengzhou (Figure 6) [56,57]. The trade-offs and synergies between the multifunctionality of arable land show significant spatial and temporal heterogeneity, which is determined by a combination of endogenous functional interactions and external drivers. Long-term observations show that the trade-off relationship between carbon stocks and nutrients in plowed soils is particularly prominent in high-intensity agricultural areas, while synergistic effects are mostly found in ecologically friendly management areas [62]. Time-series analyses show that the imbalance of arable land functions persisted in the black soil region of Northeast China from 2005 to 2020, with economic function enhancement dominating the synergistic effect and ecological function lagging behind to exacerbate the trade-offs [63]. For spatial differentiation, topography and resource endowment are key. High-slope cropland showed strong trade-offs due to degradation of soil carbon stocks and physical structure [62], and coastal areas had higher synergies between production and ecological functions, but social functions declined significantly due to urbanization. Among external factors, global warming leads to the expansion of arable land in cold regions and contraction in warm regions [64], and the EU’s Common Agricultural Policy (CAP) promotes coordination of arable land functions through ecological compensation, but an over-reliance on fertilizer subsidies may still reinforce production-ecological trade-offs. In addition, the interaction of socio-economic factors (urbanization, rural income) and natural conditions (slope, soil quality) further amplifies regional differences (Figure 7), with greater reliance on policy-adaptive management in economically developed regions [65]. Based on this study, future sensitivity analyses can be used to prioritize regulatory elements with a view to achieving optimal utilization objectives.

4.3. Policy Implications for the Sustainable Use of Cultivated Land

In the context of multifunctional management of arable land, the European Union’s Common Agricultural Policy (CAP) balances economic, environmental, and social objectives through a ‘two-pillar’ framework [65]. However, mandatory measures may lead to implementation deviations due to regional heterogeneity. The United States promotes the adoption of conservation tillage through programs such as the Environmental Quality Incentives Program (EQIP) and the Conservation Reserve Program (CRP). However, the fragmentation of these programs and their short-term contract characteristics weaken the long-term accumulation of ecological benefits [66]. Both programs attempt to enhance policy flexibility through “results-based payments” and “performance incentives”, respectively [67]. In contrast, China’s implementation of the “farmland compensation balance” and crop rotation/fallow systems faces prominent issues of unequal compensation standards between regions [68]. While rigid controls in “non-grainification” governance have safeguarded grain production capacity, the lack of a dynamic demand response mechanism has exacerbated functional conflicts in some areas [69]. Developed countries emphasize market mechanisms and cooperation among multiple stakeholders, while developing countries, constrained by fiscal and technical limitations, tend to prioritize single objectives (Figure 11). In addition, the stability of land ownership directly affects farmers’ willingness to invest in sustainable practices [70], and farmers’ behavioral decisions are influenced by multiple institutional and technical factors. In summary, in the future, it is necessary to strengthen the main position of farmers through land rights, optimize the allocation of spatial resources based on landscape and function, and use market mechanisms (e.g., ecological compensation) and policy tools (e.g., target-oriented subsidies) to guide behavioral transformation, so as to form a sustainable use model coordinated with people and the land.

4.4. Uncertainty and Prospects

The differences in the selection of indicators, modeling and parameter settings may lead to uncertainty in the results, and it is necessary to further improve the framework system in the future, so as to reveal the driving factors and influencing mechanisms at a deeper level. Restricted by the acquisition of detailed plot-level data information, this paper analyzes the situation in a single year. In the future, it is also necessary to combine remote sensing data and machine learning to cope with the problem of missing data, to carry out a multi-temporal and long time-series study, and to take into account the dynamic impact of multiple factors such as uncertain climate change, changing agricultural policies, and differentiated farmers’ behavior on the supply and demand of cultivated land functions. Setting up multi-scenario modeling and consulting stakeholders to predict the effects of the zoning strategy, will enable validation of the universality of research conclusions and the repeatability of methods.

5. Conclusions

This study proposes a framework for analyzing the sustainable use of cultivated land resources based on “landscape differentiation–functional matching”, which is used to identify the fractured space of cultivated landscapes, assess the matching of functional supply and demand, and analyze the trade-offs/synergies between sub-functions. At the landscape level, the analysis of landscape pattern yields of Zhengzhou’s arable land is dominated by the separation pattern, accounting for 77% of the total arable land area. Shared areas are mainly distributed in the intersection areas of Xinmi, Gongyi, and Dengfeng, and separated areas are mainly distributed in the border areas, with the overall characteristic of a “gradient” distribution from the center to the periphery. At the functional level, horizontally, the area with the most balanced functional supply and demand is Xingyang. Vertically, the most significant mismatch between supply and demand is between PF and EF. In terms of correlation, Cy-Si, Si-Ac, and Rbl-Frl show trade-offs, while Cy-Ac and Nei-Frl show synergies. Cultivated land resources in Zhengzhou City are divided into four first-level functional zones (PC zone, RL zone, BD zone, and MA zone), and on the basis of this, they are divided into eight second-level functional zones by combining the characteristics of the current situation. Then, corresponding use programs are formulated to promote strategic choices and elemental adjustments for the sustainable use of arable land under different circumstances.