Constructing a Composite Ecological Security Pattern Through Blind Zone Reduction and Ecological Risk Networks: A Case Study of the Middle Yangtze River Urban Agglomeration, China

Abstract

The Middle Yangtze River Urban Agglomeration, a critical ecological barrier in China, faces escalating pressures from rapid urbanization and climate change, leading to fragmented landscapes and degraded ecosystem services. To address the synergistic challenges of ecological protection and risk management, this paper takes the urban agglomeration in the middle reaches of the Yangtze River as the study area, and obtains the source patches through morphological spatial pattern analysis. Based on the spatial distribution of risky source areas, ecological blind zones are cut down by optimizing buffer zones and merging fragmented patches. Finally, a composite ecological network is constructed through circuit theory superimposed on the dual network method. The results showed that (1) there are 16 ecological source patches and 16 risk source patches in the study area. Six complementary ecological sources and four new ecological sources were obtained through the blind zone reduction strategy. The percentage of ecological blind zones reduced from 58.4% to 39.5%. (2) The integrated nodes with 11,366 connecting edges were identified. The integrated nodes are distributed around the central Jiuling-Mafushan Mountains, mainly in the western and southern areas of the Dongting Lake Plain. (3) Primary integration nodes are critical for network stability, with a 75% node failure threshold triggering systemic collapse. The proposed strategy of “mountain protection–plain control–railway monitoring” is consistent with China’s territorial and spatial planning. By incorporating the risk network into the conservation framework, this study provides feasible insights for balancing development and sustainability in ecologically fragile areas.

1. Introduction

Ecosystems are complex systems where biotic and abiotic components interact via energy–material flows within defined spatial boundaries [1]. These interactions underlie the self-regulation of ecosystems and drive the evolution of their structure and function. The network analysis approach reveals the characteristics of individual elements in an ecosystem by parsing the interactions within the ecosystem. Initially, ecological networks were primarily used to study interactions between organisms, but as landscape fragmentation and human impacts on the environment have intensified [2], the method has gradually been extended to the field of landscape ecology to address broader ecological challenges. The approach provides an important foundation for a deeper understanding of ecological processes and conservation of ecosystems and the development of appropriate conservation strategies. Network models enable multiscale simulations of energy–material flows, revealing spatial configuration principles and guiding targeted optimization of ecosystem services [3].

Currently, the research on ecological network construction has established a multiscale research system in the basic form of “source identification–resistance surface construction–corridor extraction”. With the deepening of the research on ecological network, scholars have made a lot of technical and methodological improvements from all scales, which gradually improve the ecological network construction method [4]. However, it faces a double challenge at the level of spatial layout and functional coordination. On the one hand, by constructing ecological networks, scholars are able to model the flow of matter and energy in ecosystems at different scales [5]. These studies aim to enhance ecosystem stability, maintain species richness, and coordinate conservation and development [6]. With continued human encroachment, habitat fragmentation has occurred and structural blindness has been created. To address this problem, studies have focused on optimizing ecological networks by increasing ecological sources, reconstructing ecological corridors and adding new stepping stones [7]. However, the traditional layout optimization methods rely on subjective experience and result in a limited improvement of network connectivity [8]. Recently, the concept of “ecological blind zones” [9] has been proposed to enhance the scientific and rationality of the patch selection process by identifying and supplementing key patches in ecological blind zones [10], so as to optimize the layout design of the ecological network [11]; on the other hand, since current ecological research mainly focuses on ecological protection [12], scholars often tend to study areas such as forests and grasslands [13]. These areas are ecologically superior and tend to be ecological service supply areas, and there is a strong positive ecological flow between areas. In addition, there are many ecologically sensitive areas. These areas exhibit characteristics such as high ecosystem vulnerability, low environmental carrying capacity thresholds, and weak self-recovery ability, and are unusually sensitive to external disturbances. A complex and dense negative material and energy exchange network is formed, and the negative ecological flows along the network may offset the protection effectiveness of the ecological network. However, studies of ecological risk networks tend to reduce them to a part of the resistance surface of the ecological network [14]. Neither risk transfer pathways between their interiors nor internal interactions are analyzed. The neglect of positive and negative ecological flow synergy mechanisms in existing ecological network models constrains the efficacy of functional coordination between ecological protection and risk management.

In order to formulate effective ecological conservation strategies, maintain ecosystem integrity, and promote sustainable development, it is necessary to construct a composite framework integrating positive and negative feedbacks to study the network structure of such areas. For this reason, this study further proposes a holistic composite framework covering ecological conservation and ecological risk. The concept of risk network (

Table 1. Concepts related to ecological and risk networks.

Name	Concept	Connotation	Source	Corridor	Network
Ecological network	Designed to maintain landscape integrity and species richness—a spatial entity of landscape ecological networks of nodes, corridors and patches.	The network structure system consisting of patches and corridors has become an important tool for ecosystem and species richness conservation. It also aims to optimize landscape patterns and improve landscape quality.	Basis for habitat and dispersal of living species. High ecosystem service value patches.	It is an interconnected ecological space between objectively existing ecological source sites.	Increase structural and functional linkages between the region’s fragmented and isolated ecological source sites to maintain regional patterns of ecological security.
Risk network	A type of ecostructural modeling that simulates the flow of hazardous substances and energy between ecological regions by creating negative structure-function linkages between ecologically discrete, isolated, and other regions.	Ecologically sensitive areas have a high level of ecological risk, have poor ecosystem stability, and are vulnerable to disturbance from other sensitive areas. Negative ecological flows can develop between sensitive areas.	It is a patch with poor ecosystem stability, low carrying capacity, and weak resistance to external disturbances.	It is an objective ecological space of interconnections between the territories of risk sources.	Strengthened structural-functional connectivity among fragmented risk sources exacerbates regional ecological risks while degrading systemic ecological resilience.

) was introduced to characterize the negative ecological flows between ecologically sensitive areas [15]. Risk networks, like ecological networks, contain source sites, and there are ecological flows of material circulation, energy flow, and information transfer between the various source sites, which in turn form closely connected corridor spaces. In ecological networks, these connections are usually positively correlated, which is characterized by the enhancement of ecological functions and services [16], while in risk networks, the connections are negatively correlated, reflecting the propagation and accumulation of ecological risks. Multidimensional analysis of positive–negative flow interactions enables optimal ecological source allocation and targeted risk corridor control, directly enhancing regional ecological security. Through the comprehensive study of ecological risk network, the competitive and synergistic mechanisms of ecological flows between the two networks and the comprehensive effects of their interactions can be explored in depth. This will in turn enhance the stability and adaptive capacity of the ecosystem.

To address spatial and functional deficiencies in traditional models, this study proposes a three-phase framework (blind zone reduction–ecological risk network construction–stability-driven management) for the urban agglomeration in the middle reaches of the Yangtze River. This study aims to: (1) quantify the effect of blind zone reduction and optimizing network connectivity; (2) construct an ecological risk network based on positive and negative duality and enhance functional coordination capacity; and (3) identify key nodes and hierarchically control risks.

2. Data Sources and Research Methodology

2.1. Overview of the Study Area and Data Sources

The Middle Yangtze River Urban Agglomeration is situated in central China (Figure 1), covering 314,502.7 km² across Hubei, Hunan, Jiangxi, and Anhui provinces. The region has a rich variety of landform types, including plains, hills, and low mountains, structured around the Yangtze River and its major tributaries Han, Xiang, and Gan rivers. Characterized by a subtropical monsoon climate, the area receives 1200–1800 mm annual precipitation with a mean temperature of 16–20 °C, supporting biodiversity and seasonal hydrological variations. Dominant vegetation includes evergreen broad-leaved forests and wetland hygrophytic communities. Soils transition from fertile alluvial plains to red-yellow hilly loams, sustaining rice paddies and tea plantations. The diverse topography of the region has resulted in complex ecosystems, which can be divided into forest, urban, farmland, and water ecosystems according to land use, especially with important freshwater lakes such as Dongting Lake and Poyang Lake, which are crucial to maintaining the regional ecological balance. Since 2000, however, rapid urbanization has fragmented natural habitats, and the expansion of infrastructure [17], represented by railroads, has intersected with important migration corridors, exacerbating human–wildlife conflicts.

The data of this study mainly include land use data, elevation data, basic geographic information data, etc. The data are uniformly projected to the WGS_1984 coordinate system to ensure spatial consistency, and the specific data sources are shown in

Table 2. Main data sources of the research area.

Dataset	Sources
Water	National Geomatics Center of China (http://www.ngcc.cn/, accessed on 10 November, 2024)
Road	National Geomatics Center of China (http://www.ngcc.cn/, accessed on 10 November, 2024)
Railway	National Geomatics Center of China (http://www.ngcc.cn/, accessed on 10 November, 2024)
DEM	Geographical spatial data cloud (https://www.gscloud.cn/, accessed on 11 November, 2024)
Land use	Resource and Environment Science and Data Center, Chinese Academy of Sciences (https://www.resdc.cn/, accessed on 11 November, 2024)

.

2.2. Research Methodology

In this study, a dual-objective oriented approach of blind zone reduction and an ecological risk network is proposed. The research framework diagram is shown in Figure 2.

2.2.1. Identification of Ecological and Risk Sources

Source site identification is an important prerequisite for network construction [18]. Ecological source sites are the basis for positive ecosystem effects [19]. Risk source sites [20], on the other hand, are sources of adverse ecological impacts and are usually located in ecologically sensitive areas. Distinguishing between these two types of ecological source sites can lead to better ecological protection and risk management.Specific source identification methods are shown in Table 3.

There are various methods for source site identification [21]; this study adopts the morphological spatial pattern analysis method for source site identification [22], and generates seven landscape types through the Guidos Toolbox software for morphological spatial pattern analysis [23]. According to the actual situation of the study area, the area threshold of ecological source sites and risk source sites was set at 2000 km² [24].

Table 3. Source identification methods.

Patch Type	Identification Methods
Ecological source	(1). Directly setting up nature reserves, scenic spots, or large areas of forests and waters with important ecological service values as source sites;
	(2). Landscape connectivity and morphological spatial pattern analysis methods were used to screen ecological source sites [25];
	(3). Screening of final ecological source sites through quantitative integrated assessment by evaluating the results of ecosystem service importance and ecological sensitivity in the study area [26].
Risk source	(1). Setting up areas with poor natural environments and high ecological sensitivity, such as deserts, saline and drylands, and development sites as risk source sites;
	(2). Screening of risk source sites through quantitative methods such as morpho-spatial pattern analysis and ecological sensitivity evaluation.

Table 3. Source identification methods.

Patch Type	Identification Methods
Ecological source	(1). Directly setting up nature reserves, scenic spots, or large areas of forests and waters with important ecological service values as source sites;
	(2). Landscape connectivity and morphological spatial pattern analysis methods were used to screen ecological source sites [25];
	(3). Screening of final ecological source sites through quantitative integrated assessment by evaluating the results of ecosystem service importance and ecological sensitivity in the study area [26].
Risk source	(1). Setting up areas with poor natural environments and high ecological sensitivity, such as deserts, saline and drylands, and development sites as risk source sites;
	(2). Screening of risk source sites through quantitative methods such as morpho-spatial pattern analysis and ecological sensitivity evaluation.

2.2.2. Blind Zone Reduction

Ecological blind zone refers to the failure to fully recognize and protect certain ecological processes [27], species, or ecological environments in the process of ecosystem research, protection, and management due to technical constraints, insufficient knowledge, or insufficient attention [9]. The urban agglomeration in the middle reaches of the Yangtze River is rich in water resources, and water bodies play an important role in this area, which is extremely important for the economic development and ecological balance of the urban agglomeration in the middle reaches of the Yangtze River. Therefore, the ecological function of water bodies should not be ignored when identifying ecological blind zones [13]. The reduction of blind zones mainly includes the following steps. (1) Screening core area patches smaller than or equal to the area threshold but larger than 100 km² in the original ecological blind zone, merging ecological patches with good agglomeration, and re-designating them as complementary ecological source sites to reduce the ecological blind zone. (2) To reduce the ecological blind zones by adding new ecological source sites in areas where there is a lack of ecological patches in the ecological blind zones. (3) Reduction of ecological blind zones by constructing ecological corridors between three ecological source areas. (4) Ecological networks and water bodies larger than 10 km² were analyzed for buffer zones, and the remaining core areas and water bodies larger than 100 km² were used as non-source-type patches to further reduce ecological blind zones.

Point kernel density analysis can transform discrete point data into a continuous density surface, thus revealing the aggregation characteristics of its spatial distribution and effectively identifying the location of ecologically blind areas. The following are specific formulas for kernel density analysis:

$$f\left(x,y\right)={\displaystyle \frac{1}{n\cdot h^2}}{\displaystyle \sum }_{i=1}^n\left[{\displaystyle \frac{3}{\pi }}\cdot {\left(1-{\left({\displaystyle \frac{d_i}{h}}\right)}^2\right)}^2\cdot w_i\right] \mathrm{for} d_i\le h$$

(1)

where $f\left(x,y\right)$ is the density value at location $\left(x,y\right)$ (units: points per area unit); $h$ is the bandwidth (search radius); $d_i$ is the Euclidean distance from point $i$ to $\left(x,y\right); w_i$ is the weight of point $i$.

2.2.3. Construction of Ecological Risk Networks

The resistance surface reflects the degree to which the substrate impedes the flow of matter and energy [28]. Usually, areas with higher elevations and slopes have a significant blocking effect on all kinds of material and energy flows [29]. As for land use, the ecological resistance surface and risk resistance surface of different land use types tend to show an opposite resistance value [30]. By defining different resistance factors (

Table 4. Resistance factor, resistance value grading, and weighting.

Drag Factor		Resistance Value					Weights
		1	3	5	7	9
Elevation/m	Ecological Networks	<734	734~1083	1083~1387	1387~1826	>1826	0.38
	Risk networks	<734	734~1083	1083~1387	1387~1826	>1826
Slope/°	Ecological networks	6	6~14	14~22	22~32	>32	0.11
	Risk networks	6	6~14	14~22	22~32	>32
Land use type	Ecological networks	Forested & waters	Grassland	Arable land	Unused land	Building land	0.19
	Risk networks	Building land	Unused land	Arable land	Grassland	Forested & waters
Road distance/m	Ecological networks	>5000	3500~5000	1500~3500	500~1500	<500	0.08
	Risk networks	<500	500~1500	1500~3500	3500~5000	>5000
Railway distance/m	Ecological networks	>5000	3500~5000	1500~3500	500~1500	<500	0.07
	Risk networks	<500	500~1500	1500~3500	3500~5000	>5000
Water distance/m	Ecological networks	<500	500~1500	1500~3500	3500~5000	>5000	0.08
	Risk networks	<500	500~1500	1500~3500	3500~5000	>5000
Building land distance/m	Ecological networks	>5000	3500~5000	1500~3500	500~1500	<500	0.09
	Risk networks	<500	500~1500	1500~3500	3500~5000	>5000

), it is possible to construct resistance surfaces that characterize the connectivity between network nodes [31]. Referring to the papers [32] and experts’ opinions, the resistance values of resistance factors and their weights are obtained.

Ecological corridors are optimal channels for ecological flows generated by positive ecological flows between ecological source sites [33]. Risk corridors are optimal channels for ecological flows generated by negative connectivity between risky source sites. The cost connectivity tool of ArcGIS software was used to identify areas with low resistance value between neighboring source sites. Corridors were generated by the Cost Distance tool [34]. The formula for calculating the minimum cost distance [35] is as follows:

$$C\left(p\right)=\underset{\Gamma \in \mathcal{P}}{\min }\left\{{\displaystyle \sum }_{i=1}^{n}c\left(r_i\right)\cdot d\left(r_i,r_{i+1}\right)\right\}$$

(2)

where $\mathcal{P}$ is the set of all possible paths $\mathsf{\Gamma }$ from the source to $p; c\left(r_i\right)$ represents the cost value at raster cell $r_i; d\left(r_i,r_{i+1}\right)$ denotes the distance between adjacent cells $r_i$ and $r_{i+1}; n$ is the number of cells along path $\Gamma$.

Some ecological source sites were added through the ecological blind zone reduction method above, and the ecological corridors between these ecological source sites are also worthy of consideration. To ensure the validity of the corridors [36], only the 10% with the shortest distances were selected for analysis.

In the abstraction of ecological (risk) sources as points, ecological (risk) source points are located in the center of ecological sources. Ecological (risk) corridors are used as connecting edges to form the ecological (risk) network. The two networks are superimposed to form an ecological risk network.

2.2.4. Integrated Node Identification

Nodes are critical for landscape connectivity and ecological function. Integrated nodes play a key role in maintaining the stability of the ecological network [25]. Integrated nodes are key areas where material exchange and energy flow intersect between ecologically important and ecologically sensitive areas, respectively. Therefore, an ecological approach must be adopted to improve the ecological environmental quality of each node to take protection and risk control measures to improve regional ecological security and reduce ecological risks.

Stability evaluation methods are often used to assess the structural characteristics of networks [37]. Robustness analysis can determine the stability and reliability of a network when it suffers damage [38]. The robustness, homogeneity, and network fragmentation (

Table 5. Network evaluation parameters.

Parameter	Equation	Introduction
Connectivity robustness [39]	$R={\displaystyle \frac{C}{n-n^r}}$	R refers to the connectivity robustness, indicating the connectivity condition; C represents the maximum value of the node of the largest connected subgraph in the network; n represents the number of nodes in the original network; nr represents the number of nodes deleted from the network; n − nr represents the difference in the number of nodes before and after deletion.
Global efficiency	$E={\displaystyle \frac{1}{n\left(n-1\right)}}{\displaystyle {\displaystyle \sum }_{i\ne j\in G}}{\displaystyle \frac{1}{d_{ij}}}$	E is network global efficiency; n is the number of nodes; i and j represent any different nodes in the same network that belong to the set of G nodes; dij represents the minimum distance between two nodes. In this paper, Pajek software is used to calculate global efficiency in R language.
Articulation Vertices	—	After a node in the network is deleted, the network is divided into two parts. At this time, the network fragmentation performance accurately measures the vulnerability of connections between network nodes. The higher the network fragmentation, the more vulnerable the network, and the greater the probability of decomposition after external attacks.
Assortativity	${\rho }_D={\displaystyle \frac{{{\displaystyle \sum }}_{ij}ij\left(e_{ij}-q_i{q}_j\right)}{{{\displaystyle \sum }}_i{j}^2{q}_j-{[{{\displaystyle \sum }}_{j}j{q}_j]}^2}}$	In the formula, ${\rho }_D$ is the assortativity of the network; $\mathrm{i} \mathrm{and} \mathrm{j}$ are any two nodes in the network G; $e_{ij}$ is the joint probability distribution of the residual degree of two vertices at any end of the randomly selected edge; q is the normalized distribution of residual degree.

) of the network are comprehensively analyzed to measure its stability and resilience in the face of external disturbances.

The network stability of the study area was measured by the simulation of two scenarios. (1) Scenario A: random damage scenario. Patches are subjected to random damage, and this scenario is a standard scenario for studying network stability. (2) Scenario B: human interference scenario. The lower the node level, the higher the probability of human damage to it.

3. Results

3.1. Ecological Network Construction

A total of 16 ecological source sites were identified (total area: 90,990.6 km², 28.93% of the study area), primarily distributed in mid-elevation rolling terrains surrounding the Yangtze River urban agglomeration and the south-central regions, including the Wushan Mountains in western Hubei, the Huangshan Mountain in eastern Jiangxi, and the Luoxiao-Jinggang Mountains along the Hunan-Jiangxi border. These sources are sparsely distributed in lowland plains due to intensive human activity.

High ecological resistance areas are primarily concentrated in the Dabie Mountain Residue area (northeast of Wuhan) and the Wuyi Mountain Range (south of Fuzhou) (Figure 3a), characterized by high-altitude mountainous terrains. High resistance areas are also mostly located in Wuhan, Changsha, and Nanchang, which are located in the capitals or major cities of the three provinces, with landscape land dominated by construction land, complex roads, large populations, and frequent human activities. Low-resistance value areas are located in the non-urban areas of Poyang Lake Plain, Dongting Lake Plain and Jianghan Plain. These areas have flat topography and dense river networks. As shown in the ecological network diagram (Figure 3b), 643 ecological corridors were identified. They are mainly located in the Dongting Lake Plain area, Xuefeng Mountain Range, the southern part of Jianghan Plain, Jiuling-Mufushan Mountain Range, and Luoxiao-Jinggang Mountain Range.

3.2. Blind Zone Identification and Reduction

The original ecological blind zone covers 183,780 km² (58.4% of the study area). Figure 4a shows the distribution of the original ecological blind zones, which can be seen as scattered, with large blind zones mainly concentrated in the north-central part of the Jianghan Plain and the eastern part of the Poyang Lake Plain. This spatial pattern underscores the need for urgent optimization of the initial ecological network to mitigate blind zone impacts.

Based on field conditions and expert consensus, buffer radii were established as 1500 m (ecological sources), 500 m (ecological corridors), 1500 m (water bodies), and 800 m (non-source-type patches). Buffer analysis was subsequently applied to these zones to reduce the ecological blind zone.

(1) Optimized complementary ecological sources: Six complementary ecological sources (total area: 9860 km²) were established, primarily distributed in the southern Yangtze River urban agglomeration.

(2) New ecological source integration: Four new ecological sources (radius: 3500 m) were added to non-risk zones within the original ecological blind zone, concentrated in the northern Yangtze River urban agglomeration and the Jiuling-Mufushan mountain range.

(3) Ecological corridor adjustment: A total of 769 complementary corridors were generated between ecological sources, complementary sources, and new ecological sources. New ecological sources connected only to their nearest counterparts. This reduced the ecological blind zone to 162,122.5 km² (Figure 4b), i.e., a decrease of 19,586 km². Subsequent addition of 168 new corridors further reduced the blind zone to 160,850.8 km² (Figure 4c), with an additional 1271.7 km² reduction.

(4) Non-source patch optimization: Screening and buffer analysis of 25,275.2 km² non-source patches yielded a finalized ecological blind zone III of 124,316.8 km² (Figure 4d), reducing the blind zone proportion to 39.5%.

3.3. Risk Network Construction

High altitude and steep terrain also impede the material and energy flow of the risk network. High-risk resistance areas on the risk resistance surface coincide with ecological network resistance surfaces. As shown in risk resistance surface map (Figure 5a), the low resistance areas are mostly located in the urban areas of each administrative district. Obviously, Wuhan, Changsha, and Nanchang show more blue color and have lower risk resistance. In addition, the low resistance zones are also linearly distributed along the railroad. Figure 5b shows 16 risk source locations with a land area of 98,594.9 km², accounting for 31.3% of the whole study area, mostly located in the Dongting Lake Plain, Hanjiang Plain, Poyang Lake Plain, and other areas with lower terrain. A total of 382 risk corridors were identified. The risk corridors are generally distributed along the railroads, ring-shaped around the Jiuling-Mufushan area, and mostly in the plains.

3.4. Spatial Distribution of Ecological Risk Networks

Overall in the risk network (Figure 6b), the risk areas of the urban agglomeration in the middle reaches of the Yangtze River are mainly concentrated in the plains of the middle reaches of the Yangtze River. The ecological zones are concentrated around the urban agglomerations in the middle reaches of the Yangtze River and in the mid-elevation areas. Risk regions demonstrate significant spatial interconnectivity throughout the study area. On the contrary, the ecological risk network shows obvious spatial distribution characteristics of fragmentation.

Ecological corridors overlaid with risk corridors get 401 integrated nodes. Integrated nodes are divided according to ecological corridors, the intersection of ecological corridors and risk corridors are tertiary integration nodes, the intersection of new ecological nodes and risk corridors are secondary integration nodes, and the intersection of new ecological nodes and risk corridors are tertiary integration nodes. There are 230 tertiary integration nodes, 152 secondary integration nodes, and 19 tertiary integration nodes on the integrated node distribution map (Figure 7a).

As shown in the ecological risk network map (Figure 7b), the integrated nodes are mainly concentrated in the Dongting Lake Plain area on the west side of the urban agglomeration in the middle reaches of the Yangtze River as well as in the area on the south side of the Dongting Lake Plain, where a complex and tight ecological risk network is formed, indicating that the convective effects of positive and negative ecological flows in the region are extremely significant. Therefore, the overall ecological management approach in this region should avoid the continuous deterioration of the ecological environment. In the central region of the city cluster in the middle reaches of the Yangtze River, there is a large area of ecological source sites separating the risk source sites of the three provinces, and its internal corridors are densely populated, with integrated nodes distributed linearly along the ecological blind zones and railroads.

3.5. Stability Analysis

The initial value of ecological risk network connectivity robustness is set to 1 (Figure 8). The connectivity robustness of Scenario B shows two plunge decreases when about 25% and about 75% nodes are deleted; when about 96% nodes are deleted, the connectivity robustness decreases to the lowest point successively in both scenarios.

The global efficiency starts at 0.28. Compared to the near slow and uniform decline in global efficiency under Scenario A, Scenario B shows the first plunge at 6% node deletion and the decline is much earlier.

The network fragmentation of Scenario B peaks first, and the ecological risk network under this scenario is more fragile, while the peak of Scenario A occurs later, which suggests that the ecological risk network under Scenario A is able to maintain a higher level of stability.

Ecological risk network congruence under both scenarios showed a wave-like change, with large differences in network structure. The morphology of ecological risk network under Scenario B changes faster and with a larger magnitude, which indicates that the integrated nodes with better environmental conditions are better able to maintain the stability of ecological risk network. The network congruence under Scenario B is mostly negative, which indicates that when some high-level nodes are damaged, high-level nodes are more inclined to connect to low-level nodes, and the connection between high-level nodes is cut off, which makes the network more fragile and more prone to collapse by external attacks.

This shows that the critical threshold of node failure ratio is 75%: at this ratio, the node failure network connectivity and stability plummet, and over 96% node failure leads to network collapse. In addition, although the number of nodes has a significant impact on network stability, the class of nodes also significantly affects the stability of the network.

4. Discussion

4.1. Purpose and Significance of Ecological Risk Network Research

This study develops a composite ecological security framework integrating MSPA and minimum cumulative resistance methodologies, systematically constructing an ecological risk network through four operational phases: source identification, ecological blind zone reduction, corridor identification, and network superposition. Our model advances conventional approaches by dynamically quantifying negative ecological flows, superseding the traditional static resistance surface paradigm for ecologically sensitive zones. This dual-flow analytical framework expands the theoretical scope of landscape ecology beyond unidirectional flow optimization [32], establishing new mechanistic insights through bidirectional ecological flow interaction analysis. Based on the spatial characteristics of the network, a three-level control strategy of “mountain protection, plain prevention and control, and railroad monitoring” is proposed: in mountainous areas, strict protection is implemented to enhance the regional ecological quality and system stability [40]; in plain areas, risk management is taken as the core, controlling the spread of pollution and mitigating the hedge effect of the positive and negative ecological flows in areas with intensive human activities [41]; and in railways, the systematic strategy can enhance connectivity of fragmented ecological sources and maximize the function of ecological services [20]. The systematic strategy can enhance the connectivity of fragmented ecological sources and maximize the ecological service function [42].

Beyond the integrated ecological risk network framework, both ecological and risk networks retain standalone governance utility through distinct operational mechanisms. While ecological and risk networks are both important tools for understanding and managing complex ecosystems [43], they have completely different concerns. Specifically, ecological networks focus on the maintenance of biodiversity, the establishment of species migration pathways, and the safeguarding of ecosystem services [44]. Risk networks, on the other hand, focus on identifying and quantifying risk factors in the environment, such as the spread of pollutants and areas prone to natural disasters, with the aim of minimizing negative impacts and improving environmental security. Thus, ecological networks can enhance the resilience of socio-economic systems by strengthening biodiversity corridors and ecosystem service hubs that are key to sustainable agriculture and ecotourism. Risk networks, on the other hand, mitigate economic losses and promote balanced regional development by targeting pollution control and disaster risk reduction in vulnerable areas.

Ecological risk network construction transcends physical material transfers, achieving multidimensional socio-ecological integration through functional synergy mapping, governance coupling mechanisms, and risk signaling architectures. The virtual link approach does not specify the absolute position of each agent in the virtual structure relative to a single reference system [45], but rather indicates the virtual link to which each agent belongs and its relative position within this virtual links [46]. This breakthrough from the traditional “material flow” perspective provides a more adaptive and flexible theoretical framework for ecosystem management [46].

4.2. Purpose and Significance of Blind Zone Reduction

Ecological blind zone reduction identifies and enhances historically overlooked areas in environmental regulation, establishing critical foundations for maintaining ecological network connectivity and stability [10]. Ecological blind zones, as “functionally vacant areas” not covered by existing networks, directly weaken ecosystem connectivity and service provisioning capacity [47]. In the process of constructing the ecological network, due to the existence of ecological blind zones, the landscape fragmentation is aggravated, the stability of the constructed ecological network is reduced, and the anti-interference ability is weaker. The traditional blind zone reduction strategy of adding new source sites and corridors according to human experience ignores the synergistic effect between its internal patches, and it is difficult to realize the optimization of the global layout of the network [9]. Our methodology enables systematic blind zone optimization through synergistic source identification at high-risk locations, advancing beyond traditional subjective approaches.

In this study, the reduction of blind zones is carried out only for ecological network based on the following dual considerations: first, based on the kernel density analysis of ecological risk network nodes (Figure 9b), it can be seen that the risk network naturally has low blind zones due to the high spatial agglomeration of the source area and the dependence of the diffusion path on linear infrastructure [48]; second, after spatial superposition of ecological and risk networks, in the ecological risk patterns (Figure 9a), the area of integrated blind zones in the study area is greatly compressed compared with the single ecological network, and the residual blind zones are fragmented and small in size, and have a weak impact on the overall pattern.

In addition, the blind zone reduction strategy can provide a scientific basis for patch management. The hierarchical management of the spatial distribution of ecological blind zones can provide zoning guidance for different regions. In high-density ecological blind zones, such as the Poyang Lake Plain, the fragmented small ecological patches should be integrated, and high-value areas should be screened to enhance ecosystem services and support the sustainable development of agriculture and improvement of human settlements [27]; in low-density ecological blind zones, such as the Wushan Mountain Range, the focus should be on the protection of the existing ecological corridors and the enhancement of connectivity with the isolated patches, to ensure the ecosystem stability and long-term service capacity, and to facilitate the development of a green economy, such as eco-tourism and water conservation.

4.3. Management Implications of Integrated Nodes

Integrated nodes serve indispensable functions in achieving concurrent ecological preservation and risk mitigation [49]. Since the integrated nodes are located in the core area where positive and negative ecological flows converge, these integrated nodes not only have the function of ecological protection, which is the key node to promote positive ecological flows and maintain the ecological network operation [50], but also have the double value of controlling ecological risk network, which is an important node to manage negative ecological flows [14]. Network stability assessments reveal three nodal typologies with distinct functional hierarchies.

Primary integration nodes (high robustness core nodes) are the key connecting space for positive ecological flows between ecological source areas, and also the potential corridor for negative ecological flows between risk source areas. Positive and negative ecological flows form a high-intensity hedge here constituting a two-way dynamic antagonism. Secondary integration nodes (medium robustness transition nodes) show local synergy between medium-intensity positive ecological flows and high-intensity negative ecological flows. Although the strength of positive ecological flow is weak, it can still have a partial blocking effect on negative ecological flow and maintain functional balance. Tertiary integration nodes (low robustness edge nodes) show a significant asymmetry between weak positive ecological flows and strong negative ecological flows. Negative ecological flow has a dominant isolation effect on positive ecological flow, while the reverse regulation of positive ecological flow is weak.

This research addresses limitations in conventional single-node management by developing a tiered governance framework for integrated nodes, transitioning their role from singular protection to dual “protection–isolation” functionality. Primary integration nodes implement conservation-oriented utilization, enforcing strict protection in biodiversity hotspots like the Jiuling Mountain region while facilitating sustainable agroforestry and low-impact ecotourism in peripheral zones to support local socioeconomic sustainability. Secondary integration nodes establish green infrastructure buffers featuring transportation corridor vegetation barriers, mitigating environmental stressors and lowering public health burdens in urbanized plain regions. Tertiary integration nodes deploy adaptive risk monitoring systems synthesizing multisource data for precision interventions, permitting regulated economic activities while preventing environmental degradation.

4.4. Limitations of the Study and Future Directions

While this study advances the ecological risk network theory through preliminary modeling, several limitations warrant acknowledgment. In the construction process, a lot of existing ecological network construction methods are borrowed and socio-economic factors are ignored [51], which reduces the replicability of the method and makes it difficult to set up specific management measures with reference to the actual local economic conditions [52]. Currently, the theoretical framework for ecological risk networks requires refinement, particularly regarding resistance surface weight allocation currently dependent on unvalidated expert judgment. This methodological gap is compounded by absent calibration protocols for resistance factor weighting, compromising model robustness. The absence of field validation within the Yangtze River Mid-Reach urban agglomeration creates systemic bias propagation pathways in the modeling architecture. Although the ecological network and the risk network are integrated, the dynamic interaction between positive and negative ecological flows is not quantitatively analyzed. No dynamic feedback mechanism is introduced, such as climate change and human activities, which makes it difficult to predict the long-term network trends [53].

Nevertheless, ecological risk network modeling shows great potential in dealing with complex areas with strong convective effects of positive and negative ecological flows. However, scientifically sound construction of ecological risk networks requires more in-depth exploration. The key is to ensure the comprehensiveness and scientific validity of ecological value assessment, which implies that the whole ecosystem must be analyzed exhaustively in order to accurately identify ecological sources and risk sources [54].

5. Conclusions

This study establishes an integrated ecological risk network framework incorporating blind zone reduction, achieves synergistic ecological protection and risk control through simulated bidirectional ecological flows, and offers actionable strategies for managing ecologically sensitive areas. The main conclusions are as follows. The blind zone reduction strategy effectively addresses spatial fragmentation in ecological networks while improving connectivity. A total of 16 ecological source sites and 643 ecological corridors have been identified. They are distributed in Wushan Mountain, Huangshan Mountain, Jiuling Mountain, and Mufushan Mountain. Additionally, 16 risk source sites and 382 risk corridors were delineated. Most of them are located in the Dongting Lake Plain, Hanjiang River Plain, and Poyang Lake Plain. The new ecological sources were established in addition to the risk sources, which is better than the traditional method of adding sources based on experience. Six complementary ecological sources and four New ecological sources were added through the strategy. Ecological blind zones decreased from 58.4% to 39.5% for a total reduction of 19.5%. The ecological risk network facilitates bidirectional dynamic analysis through integrated positive and negative ecological flows. The network identified 230 primary, 152 secondary, and 19 tertiary integration nodes. These nodes concentrate in the western Dongting Lake Plain region and adjacent low-elevation plains to the south. In addition, numerous primary integration nodes are distributed along the railroad. This spatial configuration supports a tiered management framework: “mountain protection–plain prevention and control–railway monitoring”. Primary integration nodes are crucial in facilitating positive ecological flows and preventing negative ecological flows, warranting prioritized conservation. Simulated anthropogenic disruption reveals catastrophic network destabilization: the global failure of primary integration nodes will trigger a precipitous decline (20% decrease) in network stability, followed by a slow decline and a rapid collapse of the network when only tertiary integration nodes are left. This hierarchical vulnerability pattern confirms descending stability contributions from primary integration nodes to tertiary integration nodes. These findings inform urgent land management protocols, including targeted reforestation and adaptive zoning regulations, to prevent ecological risk propagation.