Construction and Analysis of the Ecological Security Pattern in Territorial Space in Shaanxi of the Yellow River Basin, China

Abstract

In the context of rapid urbanization and extreme climate change globally, balancing ecological resources and economic development for land spatial planning has become one of the pressing issues that need to be addressed. This study proposes a composite model to construct a spatial ecological security pattern. It identifies restoration areas with different risk levels based on the spatial distribution of land use, offering suggestions for optimizing spatial configuration. Focusing on the central Shaanxi region of the Yellow River Basin in China, ecological sources are identified by integrating ecological factors, and ecological corridors and restoration zones are extracted using the minimum cumulative resistance difference and circuit theory. The results indicate significant improvements in ecological quality and desertification in the study area from 2000 to 2020. Currently, the core area covers 51,649.71 km², accounting for 62.18% of all landscape types; the total ecological source area covers 31,304.88 km², representing 18.84% of the entire area. These ecological source areas are mainly distributed in the northern Loess Plateau and the southern mountainous regions. The area has 26 important ecological corridors, identifying 16 ecological pinch points and 12 ecological barriers, presenting an ecological security pattern characterized by a grid-like structure in the northern region and a dispersed pattern in the southern region. Additionally, 273.72 km² of ecological restoration priority areas and 197.98 square kilometers of ecological restoration encouragement areas are proposed as key planning regions for ecological environmental protection. This study provides references for optimizing spatial configuration to promote the sustainable development of urban and rural living environments in the Yellow River Basin.

1. Introduction

The Yellow River Basin plays a vital role in supporting regional development in China [1]. In the context of rapid social and economic growth, urbanization has progressed swiftly, leading to a significant increase in urban construction land. At the same time, ecological spaces have gradually been diminished and squeezed. Consequently, a range of ecological security issues, such as the fragmentation of landscapes and the depletion of biodiversity, have become more prominent [2,3]. The ecological security framework is essential for aligning ecosystem conservation with human development. It is critical for maintaining the integrity of landscapes and ensuring the ecological safety of the region [4].

Ecological source areas are essential components in the flow of ecological elements and the provision of ecosystem services [5]. Ecological corridors function as vital connectors between these source areas, promoting the exchange of elements between ecological patches and facilitating the movement, migration, and interaction of species. These corridors not only serve critical ecological roles, such as enhancing biodiversity, purifying water, reducing soil pollutants, conserving soil and water, regulating local microclimates, and mitigating flood risks, but also act as important buffers between urban clusters. This helps curb uncontrolled urban expansion, which is key to maintaining the ecological security framework of cities [6]. Developing a regional ecological security pattern based on the “source-corridor” model involves using ecological corridors to link fragmented landscape patches across the region.

Developing a rational ecological pattern requires a comprehensive understanding of the interactions between ecosystems and landscape configurations. Common methods for this analysis include morphological spatial pattern analysis (MSPA), the minimum cumulative resistance (MCR) model, and gravity models [7,8]. The framework of “identifying ecological source areas, constructing a resistance surface, extracting ecological corridors” has become a widely accepted approach in territorial spatial planning research [9]. Different administrative units at varying scales, such as ecological protection areas [10], forest belts [11], and watersheds [12,13,14], are commonly used as research units. In response to ecological and environmental challenges in the Yellow River Basin, there has been growing academic attention on the environmental issues facing towns along the river [15,16].

Shaanxi, located in the middle reaches of the Yellow River Basin, is a key region where agriculture and animal husbandry converge [17], and it also serves as a significant energy base [18]. As the “15th Five-Year Plan” is being developed, it is essential to focus on optimizing the ecological security pattern of land use in Shaanxi. This includes identifying ecological source areas and corridors, pinpointing critical nodes for urban ecological network connectivity, determining priority zones for ecological protection and restoration, and establishing a “source-corridor” ecological security framework. These efforts will provide a scientific foundation for optimizing spatial allocation, enhancing the living environment in both urban and rural areas and fostering coordinated sustainable development between urban and regional environments.

2. Materials and Methods

2.1. Study Area

The Shaanxi region of the Yellow River Basin is situated in the central part of the basin, covering northern Shaanxi, the Guanzhong area, and parts of southern Shaanxi (33°20′–39°35′ N, 106°18′–111°15′ E; Figure 1). It is bordered to the west by the Yellow River and to the south by the Qinling Mountains. The region features diverse landforms, including plateaus, plains, and mountains stretching from north to south and experiences a typical continental monsoon climate [19]. The annual average temperature and precipitation are influenced by both latitude and topography. The northern area is mainly composed of the Loess Plateau and sandy grasslands, characterized by a warm arid climate. The central region is dominated by alluvial plains and experiences a semi-arid to semi-humid climate, while the southern area, being more mountainous, has a humid-to-semi-humid climate. The region’s overall annual average temperature is around 13.7 °C, decreasing from south to north and east to west. In January, average temperatures range from −11 °C to 3.5 °C, while in July, they range from 21 °C to 28 °C. The frost-free period spans between 160 and 250 days. From 1956 to 2023, the province’s multi-year average annual precipitation was 656.1 mm, with a notable increase in rainfall from north to south.

2.2. Data Sources

Eleven essential data points were used in this study (

Table 1. Sources of basic data.

Data	Year	Explanation	Sources
Land use type	2000, 2020, 2020	30 m land use data	Resource and Environmental Science Data Center of Chinese Academy of Sciences http://www.resdc.cn
Elevation	2019	30 m digital elevation model data	Chinese Academy of Sciences Geospatial Data Cloud http://www.gscloud.cn
Vegetation coverage	2000, 2010, 2020	30 m annual maximum NDVI dataset	National Ecosystem Science Data Center http://www.nesdc.org.cn/
Glass albedo	2000, 2010, 2020	The ratio of all reflected radiation energy to incident radiation energy in the shortwave band (0.3–3 μm)	National Earth System Science Data Centre http://www.geodata.cn
Annual average precipitation	2000–2020	Calculated from 192 precipitation stations	Shaanxi Institute of Geological Survey http://www.sxsgs.com
Annual average temperature	2000–2020	Calculated from 70 meteorological stations	Shaanxi Institute of Geological Survey http://www.sxsgs.com
slope	2019	Analyzed from DEM data	Chinese Academy of Sciences Geospatial Data Cloud http://www.gscloud.cn
GDP density	2020	GDP/land area	National Ecosystem Science Data Center http://www.nesdc.org.cn/
Distance from road	2019	Calculated from city base map data	OSM http://www.openstreetmap.org
Distance from river	2019	Calculated from city base map data	OSM http://www.openstreetmap.org
City base map and water system map	-	Including city coordinates, boundaries, railways, highways, river location information	National Platform for Common Geospatial Information Services http://www.tianditu.gov.cn/

), which met this study requirement.

3. Methodology

This study examines changes in ecological environmental quality using the MSPA-MCR model. By applying a modified resistance surface model to build the ecological security pattern of land space, the study proposes strategies for optimizing regional ecology. The detailed process is illustrated in Figure 2.

3.1. Determination of Ecological Source Areas

This study employs an overlay analysis using MSPA, landscape connectivity, habitat quality, and desertification degree to identify ecologically valuable patches, which are designated as ecological source areas within the research area.

3.1.1. MSPA Model

The MSPA model is a technique used to measure, segment, and interpret spatial patterns in raster images, focusing on spatial morphology and connectivity [20]. It accurately classifies the study area and identifies key habitat patch areas at the pixel level, making it particularly effective for recognizing ecological source areas in fragmented landscapes with dispersed ecological elements. This approach supports the scientific selection of ecological source areas. Initially, based on land use-type data, the metadata are reclassified into foreground (forest land, grassland) and background (other types). Subsequently, through image processing, the foreground is divided into seven distinct categories to highlight the connections between structures (

Table 2. Landscape types and ecological explanations based on MSPA.

Type	Ecological Explanation
Core	The habitat patches with larger pixels act as the ecological source.
Islet	Unconnected area, crumbling patches, low connection degree, less material energy exchange, and transfer.
Perforation	The edge of the internal patches in the core area.
Edge	Peripheral edge of the core area.
Loop	Ecological corridors link the same core areas and serve as corridors for species migration and material and energy flows.
Bridge	Narrow patches connect two or more core areas and serve as corridors.
Branch	Only one end is connected to the edge area, the perforation area, the bridge area, and the loop area.

).

3.1.2. Habitat Quality (HQ)

The Habitat Quality Index (HQ) is a key factor influencing ecological security patterns and is commonly used as a parameter in ecological studies [21]. In this study, the habitat quality module of InVEST 3.12 was applied to assess habitat quality. The module operates on the principle of treating certain land use types as threat factors that negatively affect the habitat system. It calculates habitat quality by considering factors such as the intensity, spatial distribution, and relative sensitivity of these threat factors. According to the model manual and the relevant literature [22], cultivated land, built-up areas, and barren land were identified as threat factors. The quality index was divided into five categories, namely very poor habitat [0, 0.2], poorer habitats [0.2, 0.4], ordinary habitat [0.4, 0.5], better habitats [0.6, 0.8], and very good habitat [0.8, 1]. As shown in Figure 3, the spatial distribution of HQ in the study area reveals considerable spatial variation, with the Guanzhong Plain and the sandy grassland areas of northern Shaanxi showing lower values compared to other regions, indicating significant interference from these influencing factors. However, significant improvements have been observed over the past decade, driven by the growing urban population’s demands for a better ecological environment.

3.1.3. Desertification

The temporal and spatial dynamics of desertification (DQ) within the study area are crucial to the ecological protection and restoration framework of the Yellow River Basin. These changes also represent a key strategic point for integrating ecological sustainability with high-quality economic development in northwestern China. The central part of the Yellow River Basin is characterized by significant desert and sandy areas, with the research area bordering the Maowusu (Mu Us) Sandy Land. Desertification significantly affects both the ecological environment and land use planning. Additionally, human activities continually drive spatial changes in desertification patterns. Therefore, the changes in desertification are a critical factor influencing the ecological sources in the study area.

Through remote sensing image correction and field exploration, a refined albedo–NDVI linear model was developed to derive the Desertification Difference Index (DDI), which quantifies the relationship between the degree of desertification and the DDI [23]. Using the density segmentation method, the DDI was categorized into five threshold ranges for the quantitative monitoring of desertification in the study area. This approach enables a more precise identification of ecological sources. The binary linear polynomial expression is as follows:

$$Albedo=-2.3744\times NDVI+1.0904$$

(1)

In the equation, −2.3744 represents the slope and 1.0904 represents theintercept of the regression equation.

A total of 2500 sample points were randomly selected for statistical regression analysis using the normalized albedo and NDVI raster data of the study area. The slope was derived from the fitted curve in the feature space, and by applying the established relationship, the final expression for the Desertification Difference Index (DDI) was determined.

The specific expression for DDI can be formulated as follows:

$$DDI=0.4212\times NDVI-Albedo$$

(2)

In the equation, 0.4212 represents the slope of the fitted line between albedo and NDVI.

Figure 4 illustrates that desertification in the study area is most prominent in the sandy grassland of northern Shaanxi and the border region of the Maowusu Sandy Land, with a notable improvement observed each year. However, as Xi’an city undergoes socio-economic development, patches of land degradation have emerged, highlighting the ongoing need for ecological restoration.

3.1.4. Landscape Connectivity Model

This model is a key indicator for assessing the connectivity between different patches, as well as material exchange, energy transfer, and species migration capacity. It plays a significant role in determining ecosystem safety [24]. In this study, based on the MSPA model, we used Conefor 2.6 to set connectivity thresholds and probabilities. By applying the integral index of connectivity (IIC), we identified patches with higher connectivity, thereby improving the accuracy of ecological source selection.

$$IIC={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n\frac{a_i\times a_j}{1+n{l}_{ij}}}}}{A_L^2}}$$

(3)

where $n$ represents the total number of patches; $a_i,a_j$ represent the areas of the patches $i,j$; $n{l}_{ij}$ represents the number of connections between patches $i,j$; and $A_L$ represents the area of the entire landscape.

3.2. Constructing the Resistance Surface Pattern

The MCR model, based on modifications to the cost distance approach [4], is widely used for constructing ecological safety patterns. The core principle of the MCR model is that the movement of biological species, nutrients, and other materials across landscapes encounters a certain level of resistance. The stronger the ecological service function and the more intact the landscape’s functions, the lower the resistance faced during these ecological processes. This model calculates the costs associated with species moving between different ecological sources, reflecting the potential patterns and trends of species movements. It effectively simulates the process of organisms navigating through various landscape matrices.

In constructing the resistance surface, various factors such as land use type, slope, the NDVI, proximity to roads and rivers, and GDP are selected as resistance factors. Drawing on relevant research [25,26] and considering the specific conditions in Shaanxi Province, resistance factor values and their corresponding weights are assigned (

Table 3. Evaluation index system of resistance factor.

Factors	Grade	Construction Resistance Score	Resistance Score	Ecological Resistance Weight
Land use type	Forest	1	1	0.31
	Grass	3	2
	Crop	5	3
	Water area	7	4
	Other settlement	9	5
Slope (°)	<8	5	1	0.15
	8~15	4	2
	15~25	3	3
	25~35	2	4
	>35	1	5
Elevation (m)	<750	1	1	0.09
	750~1100	3	2
	1100~1400	5	3
	1400~2000	7	4
	>2000	9	5
Vegetation coverage	0.80~1	1	5	0.11
	0.7~0.8	3	4
	0.6~0.7	5	3
	0.4~0.6	7	2
	<0.4	9	1
Distance from river (m)	>20,000	1	1	0.06
	10,000~20,000	3	2
	5000~10,000	5	3
	1000~5000	7	4
	<1000	9	5
Distance from river (m)	>10,000	1	1	0.02
	5000~10,000	3	2
	1500~5000	5	3
	500~1500	7	4
	<500	9	5
GDP density (CNY/km2)	<1500	1	5	0.26
	1500~2500	3	4
	2500~7000	5	3
	7000~10,000	7	2
	>10,000	9	1

). Based on relevant studies [25,26,27], the resistance factor values were determined using expert judgment, and their corresponding weights were calculated using the entropy method in SPSSPRO_1.1.29_1056 software (Table 3).

3.3. Extracting Ecological Corridors

After identifying ecological sources, the next step is to extract ecological corridors. In this study, the Linkage Mapper tool, based on weighted cost distance methods, is employed to overcome some limitations in network construction and node selection. This approach uses core habitat vector polygons and resistance rasters to map the lowest-cost connections between core areas. Each pixel in the resistance raster represents a value indicating energy expenditure, movement difficulty, or mortality risk.

The Linkage Mapper tool is used to identify ecological pinch points and barriers within ecological corridors. Ecological nodes are areas within these corridors where biological movement is concentrated, representing locations with high passage probabilities and limited alternative routes [28]. While nodes play a crucial role in maintaining connectivity, they are also at higher risk of ecological degradation. Ecological barriers, on the other hand, are areas that pose significant resistance to the movement of organisms between ecological patches. Restoring these barriers can reduce ecological resistance, thereby improving landscape connectivity between ecological sources.

4. Result

4.1. Landscape Types

Through MSPA, seven landscape categories were identified (Figure 5), and the landscape type data for various years were calculated (

Table 4. Characteristics of each landscape type.

Landscape Type	Area (km2)			In the Study Area (%)
	2000 Year	2010 Year	2020 Year	2000 Year	2010 Year	2020 Year
Core	46,825.19	52,049.39	51,649.71	58.48	62.73	62.18
Islet	691.04	587.50	631.85	0.86	0.71	0.76
Perforation	1575.24	1716.73	1742.38	1.97	2.07	2.10
Edge	14,956.47	14,539.38	14,614.45	18.68	17.52	17.60
Loop	891.62	1011.33	1061.96	1.11	1.22	1.28
Bridge	6533.67	5739.75	5892.73	8.16	6.92	7.09
Branch	8594.26	7332.47	7466.48	10.73	8.84	8.99
Total	80,067.50	82,976.55	83,059.57	100.00	100.00	100.00

). The results show that the total area of the seven landscape types in the study area is 83,059.57 km². Among these, the core area is the largest, covering 51,649.71 km², which accounts for 62.18% of the total landscape types. This represents a 3.7% increase compared to the year 2000 but a 0.55% decrease compared to 2010. The smallest landscape type is the isolated patches, which occupy only 631.85 km² or 0.76% of the total landscape types.

The gaps and edge areas between the core areas and other landscape types act as protective barriers for the core areas, comprising 2.10% and 17.60% of the total landscape types, respectively. This indicates that the core patches have relatively high stability and can effectively withstand external disturbances. The bridging areas, which function as linear corridors connecting the core areas, account for 7.09% of the total landscape and are relatively small. Compared to the year 2000, their extent has slightly decreased, and their distribution remains fragmented. This suggests that the connectivity of material flow channels between the core patches is still weak. Additionally, the area of island patches, which serve as temporary stopovers for biological migration and dispersal, remains insufficient and requires prompt expansion.

4.2. Ecological Source Areas

Considering factors such as ecosystem integrity, habitat suitability, biodiversity, material exchange, and patch fragmentation, this study identified 28 ecological source areas using MSPA, HQ, DQ, and IIC. The total area of these ecological source areas is 31,304.88 square kilometers, representing 60.61% of all ecological source patches and 18.84% of the total study area. Based on their significance, these areas are categorized into two levels, namely important ecological source areas and general ecological source areas (

Table 5. Ecological source index of 28 core patches.

No.	DIIC	HQ	DQ	Area (km2)	No.	DIIC	HQ	DQ	Area (km2)
1	2.546	0.005	0.646	479.74	15	34.985	0.232	0.800	6599.10
2	0.595	0.098	0.852	112.09	16	0.938	0.166	0.699	176.95
3	2.172	−0.059	0.793	409.24	17	2.701	0.231	0.855	509.53
4	0.989	0.058	0.731	186.29	18	2.810	0.227	0.718	530.09
5	0.922	−0.065	0.853	173.81	19	5.989	0.242	0.813	1129.69
6	1.599	−0.060	0.609	301.33	20	0.802	0.264	0.804	151.24
7	0.374	0.126	0.863	102.65	21	1.983	0.247	0.723	374.07
8	0.553	0.103	0.817	104.15	22	0.783	0.261	0.730	147.75
9	1.570	0.173	0.783	296.06	23	1.281	0.261	0.814	241.65
10	0.884	0.197	0.737	166.75	24	1.293	0.256	0.784	243.89
11	34.163	0.237	0.830	6444.17	25	56.320	0.242	0.849	10,623.50
12	0.796	0.223	0.698	150.20	26	4.868	0.228	0.718	919.50
13	0.557	0.225	0.612	105.08	27	1.170	0.250	0.816	220.71
14	1.442	0.085	0.662	272.05	28	0.701	0.267	0.821	133.60

, Figure 6).

In terms of the spatial distribution of ecological source areas, the important ecological source areas are the most prevalent, mainly concentrated in regions such as Fu County, Ganquan County, Yichuan County, Feng County, Taiba County, Zhouzhi District, Huyi District, and Tongguan County. These areas include significant ecological landscapes such as Laoshan Forest Park, Huangdi Mausoleum Scenic Area, Hukou Waterfall Scenic Area, Tiantai Mountain Scenic Area, Taibai Mountain Forest Park, Louguantai Forest Park, and Taiping Forest Park. Additionally, other ecological source areas are sparsely distributed in the northern Kuye River and Wuding River basins, as well as the Jinghe and Weihe River basins in the central urban area. Ecological source areas are relatively rare in other regions. This uneven distribution highlights the need for strengthened conservation and restoration efforts in certain areas to improve the overall stability and biodiversity of the ecosystem.

Human activities continue to have a significant impact on the ecological space within the ecological source areas of the Shaanxi section of the Yellow River Basin, resulting in the compression and encroachment of these spaces. Therefore, enhancing the protection of ecological spaces is crucial. In the process of constructing a national land ecological security pattern, more emphasis should be placed on safeguarding key ecological source areas. Additionally, it is important to focus on social equity, ensuring the protection of ecological source areas across all levels.

4.3. Resistance Surface Pattern

A weighted overlay analysis was conducted based on the resistance surfaces of various factors to generate the comprehensive resistance surface results (Figure 7). The resistance intensity in the study area is generally low, which facilitates species diffusion and migration, as well as the flow of elements. However, due to the significant variation in land types throughout the study area, the distribution of resistance values is uneven. High resistance values are predominantly found in the urban areas of various districts and in the sandy grassland regions of northern Shaanxi, with the highest resistance values concentrated in the urban area of Xi’an, located in the southern part of the study area.

4.4. Ecological Corridor

Based on the results from the Linkage Mapper tool, 26 ecological corridors were identified (Figure 8), with a total length of 1757.4 km. Among these, 21 corridors exceed 5 km in length and 16 stretch beyond 10 km. In terms of spatial distribution, the ecological corridors successfully connect ecological source areas at various levels. They are evenly spread across both northern and southern Shaanxi, significantly improving social equity in terms of human access to ecosystem services.

By constructing ecological corridors, 16 ecological pinch points (Figure 8a) and 12 ecological barrier points (Figure 8b) have been identified. These points are primarily located along the circular ecological corridors linking northern Shaanxi with the Shangluo region in southern Shaanxi. These corridors connect key ecological source areas and critical habitats essential for species survival. As areas with frequent ecological processes, they should be prioritized for protection. Additionally, ecological barrier points should be targeted for restoration efforts to facilitate the effective dispersal and migration of species between ecological source areas through these corridors.

According to existing research [29], to support species movement and biodiversity within ecological corridors, the widths of 12 m, 60 m, 200 m, and 600 m correspond to the minimum width, improved width, suitable width, and sufficient width, respectively, for protecting species movement and biodiversity. To further assess the impact of human activities on habitats along these ecological corridors, land use types within the width ranges of 12 m, 60 m, 200 m, and 600 m were analyzed (

Table 6. Land use types in ecological corridors of different widths.

Land Use Types	Width of Ecological Corridors (km2)
	12 m	60 m	200 m	600 m
Cropland	11.68	59.1	212.53	750.36
Forest land	2.99	15.06	53.52	16.95
Grassland	25.4	126.4	400.12	1044.59
Water body	0.24	1.18	4.6	20.21
Built-up land	0.39	2.02	7.6	29.9
Barren land	1.48	7.5	28.76	122.64
Total	42.18	211.26	707.13	1984.65

and Figure 8).

From the perspective of land use proportions (Table 4 and Figure 9), the distribution of forest and grassland within ecological corridors of different widths remains fairly consistent, ranging from 64% to 67%. The proportions of constructed and agricultural land vary between 28% and 39%, with this proportion increasing as corridor width expands. This suggests that the current ecological corridors are adequately designed to support species movement and biodiversity, and the development of the ecological safety pattern is progressing well. However, as the corridor width continues to increase, the proportion of constructed land also rises significantly. At a width of 600 m, constructed and agricultural land has increased by 39% compared to the original baseline. This highlights the considerable impact of human activities on habitats surrounding ecological corridors, emphasizing the need to ensure that the ecosystem retains a suitable environment and sufficient space for future development.

4.5. Identifying Key Areas for Ecological Restoration

In recent years, rapid land space utilization has led to significant investments in ecological restoration. To further enhance the connectivity of the ecological network and optimize the spatial ecological security pattern, we take into account factors such as the land use conditions of biological corridors, ecological pinch points, ecological barrier points, proximity to ecological source areas, and the difficulty of species movement. This comprehensive approach allows for the identification of key areas for ecological restoration aimed at improving the connectivity of ecological source areas and facilitating species movement through ecological corridors. Based on the requirements for ensuring species migration, preserving biodiversity, and strengthening corridor connectivity, we categorize these areas into two levels, namely priority areas for ecological restoration and encouraged areas for ecological restoration (

Table 7. The criteria for classifying key areas for ecological restoration.

Level	Define the Scope	Basis
Priority areas for ecological restoration	Construction land and agricultural land within the ecological corridor range	Construction land and agricultural land within the ecological barrier zone
Encouragement areas for ecological restoration	Construction land and agricultural land within the ecological barrier zone	Can effectively enhance the connectivity of ecological sources and the permeability of ecological corridors

). This classification serves as a basis for formulating plans to remove construction and agricultural land within ecological corridors.

Based on the identification results shown in Figure 10, 16 priority ecological restoration areas have been delineated, covering a total area of 273.72 km², along with 12 ecological restoration encouragement areas covering 197.98 km². These areas are primarily concentrated within the circular ecological corridor connecting Dingbian, Jingbian, Hengshan, Yulin, Suide, Ansai, and Yan’an, as well as the ecological source areas of Shanyang and Danfeng, which link to external corridors.

For the priority ecological restoration areas, it is essential to proactively integrate them into the upcoming round of national land spatial planning. This involves closely monitoring the ecological environment and controlling high-intensity development activities, such as production construction, mining rights land use, and the establishment of artificial commodity forests. These measures will help ensure that species can freely move between ecological source areas, thereby enhancing biodiversity. For the ecological restoration encouragement areas, it is important to gradually address the unfavorable land use patterns within the ecological framework as part of a long-term strategy. For example, after ensuring stable living conditions for rural communities, residential areas should be adjusted to protect ecological integrity. Additionally, the legitimate rights of mining rights holders and commodity forest owners should be respected, ensuring that their activities are not excessively restricted by ecological protection boundaries. This should include the phased withdrawal of mining rights and artificial commodity forest land, striking a balance between ecological preservation and socio-economic development.

5. Discussion

The research on the ecological security pattern in the Shaanxi region of the Yellow River Basin is a critical element of regional ecological governance and a crucial support for achieving high-quality development and ensuring national ecological security.

From 2000 to 2020, the ecological quality of the Yellow River Basin in Shaanxi has significantly improved, particularly in the central region. This improvement is evidenced by the reduction in large degraded areas and the partial recovery of desertified land in the northwest, aligning with China’s national strategies for ecological civilization construction and desertification control [30]. These positive trends are likely the result of targeted policies, including the Green Program [31], large-scale afforestation projects [32], and the stricter enforcement of land use regulations under the Ecological Protection Red Line policy [33].

However, regional disparities persist, particularly in the northern areas, where slower recovery can be attributed to arid climatic conditions [34] and intensive human activities such as mining and urbanization [35]. This highlights the complex interaction between policy effectiveness and the biophysical limitations of dryland ecosystems. The uneven distribution of ecological sources and the reduced connectivity among them point to the need for more comprehensive strategies to protect and enhance biodiversity.

Among the various landscape types, the core areas are the largest, while the islets are the smallest. Ecological sources are predominantly concentrated in counties such as Fuxian, Huangling, and Taibai, where they form continuous patches that are crucial for maintaining biodiversity. However, the proportion of ecological sources in this region remains lower compared to other parts of the Yellow River Basin [15,36], raising concerns about the long-term ecological sustainability of the region. The reduced connectivity between these sources exacerbates the problem, underscoring the need for the expansion of protected areas and the establishment of buffer zones to mitigate edge effects.

The integration of morphological spatial pattern analysis (MSPA) and the minimum resistance cost (MCR) model provides a robust framework for quantifying ecological sources and corridors in heterogeneous landscapes. MSPA effectively distinguishes core habitats from fragmented patches [20], while MCR optimizes the identification of ecological corridors by incorporating resistance factors such as land use, topography, and human disturbance [37]. This integrated approach aligns with recent studies emphasizing the importance of structural and functional connectivity in the Yellow River Basin [38].

Despite its advantages, this approach has certain limitations. MSPA’s reliance on binary land cover data may oversimplify ecological gradients, potentially leading to inaccurate representations of ecological diversity. Future research could improve the precision of ecological security pattern optimization by integrating multi-scale remote sensing data and developing dynamic resistance surface modeling. This would enhance the accuracy of modeling ecological changes and species interactions, advancing the understanding of how to optimize ecological security patterns in complex river basin systems.

6. Conclusions

This study systematically addresses critical scientific challenges in enhancing ecological quality and reinforcing land use security within the Yellow River Basin by employing remote sensing technology. The key findings are summarized as follows:

Over the past two decades, the ecological environment in the Shaanxi region of the Yellow River Basin has significantly improved, with noticeable reversals in desertification. According to the latest landscape classification, the core area now accounts for 62.18% of all landscape types, covering an area of 51,649.71 km².

The identified ecological source areas in the study region total 31,304.88 km², representing 18.84% of the total area. These sources are primarily concentrated in the northern Loess Plateau and the southern mountainous regions. The region features 26 key ecological corridors, with 16 ecological pinch points and 12 ecological barriers. The resulting ecological security pattern presents a network structure in the northern part and a more fragmented pattern in the southern part.

Based on this ecological security pattern, 273.72 km² of ecological restoration priority areas and 197.98 km² of ecological restoration encouragement areas have been proposed as key planning zones for ecological protection.