Assessing Ecological Security Pattern by Integrating Multiple Risks in the Structure–Process–Function Framework: A Case Study from Heilongjiang Province, China

Abstract

Ecological security is fundamental to human survival and sustainable development. However, current assessment frameworks often lack regional adaptability. They also frequently overlook the security risks associated with landscape patterns. To address these gaps, this study establishes a “structure–process–function” framework that integrates potential ecological risks. Using Heilongjiang Province as a case study, we assessed the spatiotemporal evolution of its ecological security pattern. The results indicate that: (1) The average ecological security pattern index (ESPI) values for Heilongjiang in 2000, 2010, and 2020 were 0.6869, 0.6573, and 0.6752, respectively. This trend exhibits an initial decline followed by partial recovery, with unsafe areas distributed sporadically. (2) The spatial pattern remained relatively stable but showed significant regional heterogeneity. Yichun City achieved the highest security level, while insecure areas were primarily concentrated in Daqing City. (3) Regarding specific dimensions, habitat fragmentation (structural security) was prevalent in central agricultural and urban areas; soil erosion (process security) remained significant in the southeastern regions, despite overall stability; and ecosystem services (functional security) followed a “degradation–adjustment–recovery” trend from 2000 to 2020. This framework effectively achieves the coupled assessment of landscape patterns and ecological security, providing scientific support for regional ecological management.

1. Introduction

In recent years, the world has experienced rapid urbanization [1]. Accelerated socioeconomic development and mounting anthropogenic pressures have precipitated a series of ecological and environmental issues, such as soil erosion, environmental pollution, and biodiversity loss [2,3,4]. These challenges undermine the health and resilience of ecosystems, posing severe threats to socioeconomic development and human survival [5,6,7]. Against this backdrop, the concept of ecological security was first proposed by the International Institute for Applied Systems Analysis (IIASA) in 1989 [8]. Ecological security refers to the capacity of an ecosystem to maintain a stable structure and intact functions, thereby sustaining ecosystem services for human survival and socioeconomic development while safeguarding the human living environment from external threats [9,10]. With increasing global attention, ecological security has emerged as an important issue facing humanity in the 21st century [11,12], and has become a focal point of research in disciplines including geography, ecology, and environmental science [13]. In China, ecological security has been elevated to an integral component of national security [14,15]. Actively promoting the construction of ecological civilization and optimizing the ecological security barrier system have become critical national strategies [16,17]. Recently introduced policies, including the Ecological Conservation Redline and the National Territorial Spatial Planning (2016–2030), explicitly require the precise identification of ecological risks and the optimization of protection patterns. Consequently, ensuring ecological security has become a fundamental prerequisite for meeting these strategic requirements.

Ecological security assessment serves as the foundation of ecological security research [18]. Its objective is to employ scientific and systematic methods to quantitatively analyze the security status of ecosystems, thereby providing a basis for ecological management and policy formulation [19,20]. The key step of ecological security assessment is to construct scientific evaluation standards and an index system. The earliest applied model was the pressure-state-response (PSR) framework, developed by the OECD (Organization for Economic Co-operation and Development) in 1993 [21]. This model effectively reflects the relationship between humans and the ecological environment and has been widely applied [22]. Subsequently, numerous derivative models were developed, such as DPSIR (driver–pressure–state–impact–response), DSR (driver–state–response), and DPSER (driver–pressure–state–exposure–response) [20]. Another prevalent approach involves ecological models, primarily the ecological footprint method [23], which assesses the balance between human demand and ecological carrying capacity by measuring the utilization of natural resources and products [24,25].

To safeguard ecological security, China has implemented a series of initiatives, such as the National Periodic Remote Sensing Survey and Assessment of Ecological Status and the Measures for Regional Ecological Quality Evaluation (Trial), both of which prioritize the assessment of ecological patterns. As the physical carrier of ecological processes, landscape patterns not only determine the structural integrity of ecosystems but also maintain critical ecosystem services such as soil conservation and climate regulation [26,27]. Consequently, accurate pattern assessment is a prerequisite for optimizing spatial configuration and enhancing ecosystem functions. Prevalent assessment methodologies largely rely on landscape indices, which describe landscape patterns to elucidate the link between structure and process, thereby enhancing the understanding of landscape functions [28,29]. For instance, Hu et al. evaluated the ecological pattern of Henan Province using the PSR framework and landscape indices [30]. Other studies employ the remote sensing ecological index (RSEI) for rapid ecological monitoring [31]. This index integrates four indicators essential to human survival: greenness, wetness, dryness, and heat [32]. For example, Li et al. utilized RSEI to evaluate wetland ecological patterns [33]. Currently, morphological spatial pattern analysis (MSPA) is widely adopted. Based on mathematical morphology theory [34], MSPA categorizes landscape patterns into seven classes based on local context: core, islet, edge, perforation, bridge, loop, and branch [35,36]. This method quantifies the morphology, structure, and spatial configuration of geographic landscapes, facilitating a better understanding of ecosystem composition and function [37], and focusing on structural characterization [38].

Ecological security pattern (ESP) refer to potential spatial configurations within a landscape, comprising critical strategic points and specific spatial relationships, which play a pivotal role in safeguarding and controlling essential ecological processes [39,40]. They have become an important planning tool for maintaining ecological security and promoting sustainable development [41]. Currently, research on ESPs predominantly focuses on pattern construction and optimization, establishing a widely accepted paradigm of “source identification–resistance surface construction–corridor extraction” [42,43]. Source identification constitutes the core component of this paradigm and generally falls into two categories: direct designation and evaluation-based identification. Early approaches favored direct designation, selecting areas with superior natural environments—such as nature reserves, forest parks, and stable vegetation patches with high coverage [44]. Subsequently, methodologies evolved toward extracting high-quality patches via comprehensive evaluation index systems, incorporating multidimensional definitions of ecological security such as ecosystem services, health, and sensitivity. For instance, Zhong et al. [45] determined ecological sources based on a combined assessment of ecosystem services and ecological health, while Li et al. [13] developed an importance–sensitivity–connectivity framework for source identification.

Despite these methodological advancements, several critical deficiencies persist in current research. First, in the research of ecological security assessment, indicator selection often lacks consistency. Scholars frequently disagree on interpretation and selection logic. Furthermore, the scientific suitability of specific indicators is often overlooked, leading to mechanical application or misuse in practice [46]. Additionally, methods like the ecological footprint tend to oversimplify complex ecosystem characteristics [47]. These approaches often fail to integrate specific regional conditions, which limits their practical value for local management. Second, existing landscape pattern analyses prioritize morphological description [38]. They often overlook the intrinsic security implications, such as the capacity to resist disturbances and sustain key ecological processes. Consequently, static assessment outcomes frequently fail to reflect actual dynamic risks. Third, research on ecological security patterns (ESP) tends to “prioritize construction over evaluation”. There is often no systematic framework to validate the quality of the constructed patterns. This absence limits both the scientific rigor and the practical effectiveness of ESP construction.

To address these gaps, this study proposes a novel assessment framework. This framework explicitly integrates potential regional ecological risks into a “structure–process–function” dimension. We applied this framework to Heilongjiang Province. Based on local ecological characteristics, we selected three targeted risk indicators: habitat fragmentation (Structure), soil erosion (Process), and ecosystem service degradation (Function). This approach addresses common deficiencies in the literature, particularly the lack of regional adaptability and risk orientation. Specifically, this research focuses on two goals: (1) to develop a regionally specific ecological security pattern assessment framework that incorporates risk elements; and (2) to evaluate and validate the security of landscape patterns in Heilongjiang Province, subsequently analyzing their spatiotemporal trends. Ultimately, this study aim to deepen the understanding of the pattern–security coupling mechanism. It provides a novel theoretical perspective for ecological security evaluation systems. Furthermore, it offers a robust scientific basis for the ecological security and sustainable management in Heilongjiang Province.

2. Materials and Methods

2.1. Research Area and Data Sources

Heilongjiang Province is located in northeastern China (Figure 1a). Its landscape is characterized by a composition traditionally described as five parts mountains, one part water, one part grassland, and three parts farmland, with elevations ranging from 26 to 1648 m (Figure 1b). As a key ecological security barrier in northern China, Heilongjiang Province also serves as one of the most important commodity grain production bases in the nation. The province is endowed with abundant and diverse ecological resources. Forest coverage reaches 46.14%, with forest land constituting the dominant land use type (Figure 1c). Farmland is concentrated primarily in the Songnen Plain and the Sanjiang Plain. The province accounts for 56.1% of the country’s typical black soil cultivated land, and its wetland area ranks among the highest in the nation [48,49]. Grasslands are distributed mainly in the western Songnen Plain and the hilly transition zone. However, driven by rapid urbanization and agricultural intensification, the region is increasingly subject to multiple ecological risks, including black soil degradation, wetland shrinkage, and biodiversity loss [50]. Consequently, regional ecological security is under significant threat. Therefore, a systematic assessment and optimization of the ecological pattern are urgently needed.

Figure 1. Overview of the study area: (a) geographical location, (b) digital elevation model (DEM), and (c) land use types.

This study utilized multi-source spatial data to support a comprehensive assessment and analysis of ecological security patterns in Heilongjiang Province. Detailed specifications regarding the acquisition year, spatial resolution, and source for each dataset are provided in Table 1. To ensure spatial consistency, all datasets were projected into a unified coordinate reference system (WGS_1984_UTM_Zone_51N) and resampled to a uniform spatial resolution using ArcGIS 10.8. Additionally, the calculation of landscape indices was performed using Fragstats 4.2.

Table 1. Detailed description of data used in the study.

Data	Year	Spatial Resolution	Sources
DEM	2020	1 km	http://www.resdc.cn/
Land Use	2000, 2010, 2020	30 m	http://www.resdc.cn/
Administrative Boundary	2024	-	https://www.tianditu.gov.cn/
NPP	2000, 2010, 2020	30 m	https://www.scidb.cn/
Soil Conservation Capacity	2000, 2010, 2018	300 m	https://www.scidb.cn/
Habitat for Key Protected Wildlife	2000, 2010, 2020	30 m	https://www.scidb.cn/
CHEQ	2001, 2010, 2020	500 m	http://www.geodata.cn/

2.2. Establishment of the Evaluation System

This study integrates regional characteristics within a structure–process–function framework to assess the ecological security pattern of Heilongjiang Province. Specifically, this research employs a cross-sectional analysis across three nodes (2000, 2010, and 2020) within a 21-year period. The research process is organized into four stages (Figure 2): (1) identification of ecological resources and potential ecological risks in Heilongjiang; (2) construction of a comprehensive index system for ecological security pattern assessment; (3) quantitative assessment and spatial mapping of the ecological security pattern; and (4) validation of the assessment results, coupled with analyses of spatiotemporal dynamics and comparisons across different periods.

Figure 2. Research framework for this study.

This study adopts an ecological risk perspective and integrates risk assessment within the structure–process–function framework. Guided by the policy requirements and ecological resources of Heilongjiang Province, we selected habitat fragmentation (representing structural security), soil erosion (representing process security), and ecosystem service degradation (representing functional security) as the criteria layers. Subsequently, landscape pattern factors influencing these three risks were selected as specific metrics for the indicator layer.

Landscape metrics are critical for analyzing the interaction between landscape patterns and ecological processes, capturing dimensions such as edge, shape, aggregation, and diversity [51]. Based on relevant literature [52,53,54,55,56,57,58], we selected nine indices to assess the targeted ecological risks: Patch Density (PD), Contagion Index (CONTAG), Modified Shannon’s Evenness Index (MSIEI), Largest Patch Index (LPI), Number of Patches (NP), Area-weighted Mean Patch Size (MAI), Landscape Shape Index (LSI), Aggregation Index (AI), and Shannon Diversity Index (SHDI). The definitions and ecological significance of these metrics are detailed in Table 2.

Table 2. Selection of landscape pattern index and ecological significance.

Indicator Name	Abbreviation	Ecological Significance
Patch density	PD	Reflects the degree of landscape fragmentation. Higher values indicate a more fragmented landscape and greater human disturbance.
Landscape contagion index	CONTAG	Describes the degree of aggregation and extension of patch types. High values indicate dominant patch types with good connectivity.
Modified Shannon’s Evenness Index	MSIEI	Quantifies the evenness of the distribution of different patch types. Higher values indicate a more equal distribution of area among patch types.
Largest patch index	LPI	Represents the percentage of the landscape comprising the largest patch.
Number of patches	NP	Measures the total number of patches in the landscape.
Area-weighted mean patch size	MAI	Reflects the degree of fragmentation. Smaller values suggest a more broken landscape structure with higher division tendencies.
Landscape shape index	LSI	Characterizes the geometric complexity of patch shapes. Higher deviations from a standardized shape indicate more irregular boundaries.
Aggregation index	AI	Measures the degree of clumpiness or non-randomness of patch types.
Shannon diversity index	SHDI	Indicates landscape diversity and heterogeneity. Higher values represent a richer variety of land use types and a more complex landscape composition.

2.3. Calculation of the Comprehensive Index for Ecological Security Pattern Assessment

2.3.1. Index Standardization

To address the dimensional heterogeneity among indicators, we adopted the min-max normalization method to normalize the data. This method transforms original values into a [0, 1] range to eliminate unit and magnitude differences, ensuring comparability and enabling effective weighting of diverse indicators. The equations are as follows:

For positive indicators:

$$Y_{ij}={\displaystyle \frac{X_{ij}-X_{min}}{X_{max}-X_{min}}}$$

(1)

For negative indicators:

$$Y_{ij}={\displaystyle \frac{X_{max}-X_{ij}}{X_{max}-X_{min}}}$$

(2)

where Y_ij represents the standardized value of each index; X_ij represents the original value of each index; X_max represents the maximum value; X_min represents the minimum value.

2.3.2. Calculation of Index Weights

The common methods of determining indicator weights include the analytic hierarchy process (AHP), the entropy weight method (EWM), and principal component analysis (PCA). Among these, the EWM offers a significant advantage by mitigating human interference, thus ensuring a more objective evaluation [59]. Therefore, we used EWM to calculate the weights of each landscape indicator (Table 3). The specific calculation steps are as follows:

Table 3. Ecological security pattern evaluation indicator system.

	Dimension	Risk	Indicator	Weight	Attribute
Ecological security pattern assessment	Structure	Habitat Fragmentation	PD	0.04	negative
			CONTAG	0.05	positive
			MSIEI	0.18	negative
	Process	Soil Erosion	LPI	0.28	positive
			NP	0.04	negative
			MAI	0.04	negative
	Function	Ecosystem Services Degradation	LSI	0.06	negative
			AI	0.05	positive
			SHDI	0.26	negative

Calculate the proportion of the i-th sample value to the j-th indicator (P_ij):

$$P_{ij}={\displaystyle \frac{x_{ij}}{\sum _{i=1}^n{x}_{ij}}}$$

(3)

Calculate the entropy value of the j-th indicator (e_ij):

$$e_{ij}=-k\sum _{i=1}^n{P}_{ij}\ln \left(P_{ij}\right)$$

(4)

where k = 1/ln(n), and n is the total number of samples.

Calculate the information entropy redundancy (d_i):

$$dj=1-ej$$

(5)

Calculate the weight of each indicator (w_i):

$$w_{ij}={\displaystyle \frac{d_{ij}}{\sum _{j=1}^m{d}_j}}$$

(6)

2.3.3. Calculation of the Comprehensive Ecological Security Pattern Index

The Ecological Security Pattern Index (ESPI) was calculated through the weighted summation of the indicator layer. The calculation formula is as follows:

$$\mathrm{E}\mathrm{S}\mathrm{P}\mathrm{I}={\sum }_1^n{W}_i{X}_{ij}$$

(7)

where W_i represents the weight value of each indicator; X_ij represents the standardized value of each indicator; n is the number of indicators.

2.3.4. Classification of Assessment Results

Based on relevant literature [60,61,62,63], we adopted the equidistant method to categorize the ecological security assessment of Heilongjiang Province into five levels (Table 4). The ESPI values range from 0 to 1. Higher values denote a higher level of ecological security, while lower values indicate a poorer ecological security status.

Table 4. Corresponding grading of ecological security and ecosystem service values. Regional ecological security pattern assessment standards and levels.

Ecological Security Level	Number	Standards
Very insecure	1	≤0.20
Relatively insecure	2	0.20–0.40
Moderately secure	3	0.40–0.60
Relatively secure	4	0.60–0.80
Very secure	5	≥0.80

2.4. Validation of Assessment Results

To ensure the scientific rigor and reliability of the results, this study employed four core indicators for cross-validation: net primary productivity (NPP), soil conservation capacity, habitat quality for key protected wildlife, and the China ecological quality index (CHEQ). A Spearman correlation analysis was performed to test these associations, with statistical significance determined at p < 0.01.

3. Results

3.1. Spatial Distribution and Changes of Ecological Security Pattern Assessment Results

Utilizing the assessment model, we calculated ESPI values to analyze spatiotemporal variations (Figure 3). The results indicate that the mean ESPI decreased from 0.6869 in 2000 to 0.6573 in 2010, before recovering partially to 0.6752 in 2020. Despite the recovery observed in 2020, the index remained below initial levels. Spatially, insecure zones in the southwestern region expanded significantly in 2010, while the central and eastern regions also experienced ecological deterioration. By 2020, although the overall ecological security situation had improved, the southwestern region remained a high-risk area.

Figure 3. Spatiotemporal evolution of the ESPI in Heilongjiang Province from 2000 to 2020.

Overall, from 2000 to 2020, the ESPI of Heilongjiang Province was characterized by a trajectory of initial decline followed by partial recovery. Geographically, very secure and relatively secure areas were predominantly located in the Greater Khingan Mountains, the Lesser Khingan Mountains, and the Changbai Mountains in the southeast. Moderately secure areas were concentrated in the plains, while insecure areas were distributed sporadically.

Significant changes were observed in the areal proportions of different ecological security levels in Heilongjiang Province from 2000 to 2020 (Figure 4). Overall, the relatively secure category consistently accounted for the largest proportion, maintaining a dominant and stable position throughout the study period.

Figure 4. Variations in the area percentage of ecological security categories from 2000 to 2020. (a) Temporal variation in the proportion of different ecological security levels; (b) Sankey diagram illustrating the dynamic transfer directions and flow magnitudes between security levels from 2000 to 2020.

Regarding specific phases, the areal proportion of the very secure category exhibited a trend of decline followed by recovery, decreasing significantly from 23.41% in 2000 to 17.17% in 2010 before rebounding to 20.72% in 2020. Conversely, the moderately secure category showed an inverse trend, rising from 26.01% in 2000 to 32.86% in 2010, and subsequently decreasing to 28.34% in 2020. In contrast, the coverage of relatively insecure and very insecure areas remained consistently low, with their combined proportion never exceeding 5%.

The sankey diagram analysis reveals that transitions occurred predominantly between adjacent security levels. The period 2000–2010 was characterized by evident degradation, primarily involving transitions from very secure to relatively secure or moderately secure levels. Conversely, the 2010–2020 period exhibited a marked recovery trend, with substantial transfers from moderately secure to relatively secure and very secure levels. This shift underscores a comprehensive improvement in regional ecological security, highlighting the success of recent conservation interventions.

3.2. Assessment Results at the Municipal Scale

From 2000 to 2020, the ecological security levels of prefecture-level cities in Heilongjiang Province exhibited significant spatial heterogeneity and dynamic variations (Figure 5). While the overall spatial pattern remained relatively stable, pronounced disparities in ecological security status were observed across administrative regions. High-security zones were consistently concentrated in the forested mountainous areas of the north and southeast, whereas the central plains were dominated by low-security zones.

Figure 5. Spatial distribution of ESPI levels among prefecture-level cities in Heilongjiang Province. (a) 2000, (b) 2010, (c) 2020, and (d) administrative map. (1: Mudanjiang; 2: Qiqihar; 3: Suihua; 4: Jixi; 5: Heihe; 6: Harbin; 7: Daqing; 8: Daxing’anling; 9: Yichun; 10: Hegang; 11: Shuangyashan; 12: Jiamusi; 13: Qitaihe).

Specifically, Yichun maintained the highest ecological security level, largely attributable to its extensive forest coverage and minimal anthropogenic disturbance. Heihe and Daqing were characterized predominantly by moderately insecure status. Notably, insecure areas were primarily clustered within Daqing City. This phenomenon likely stems from the adverse ecological impacts of long-term oil extraction, including habitat fragmentation, soil degradation, and vegetation destruction. The Greater Khingan region exhibited substantial fluctuations associated with forest landscape fragmentation induced by historical forestry activities. In other cities, the ecological security level remained consistently dominated by the relatively secure category, maintaining a stable spatial pattern.

3.3. Assessment Results and Changes of Different Dimensions

3.3.1. Structure Security and Habitat Fragmentation Risk

As shown in Figure 6, we analyzed the dynamics of structural ecological security (represented by habitat fragmentation risk) utilizing normalized PD, MSIEI, and CONTAG indices. Following normalization, higher values for all indicators indicate a higher level of structural ecological security and a correspondingly lower risk of habitat fragmentation.

Figure 6. Spatial patterns of structural security characterized by habitat fragmentation risks.

The PD metric indicated a decreasing trend in fragmentation, with low-fragmentation areas expanding in the central and southern regions in 2010 and 2020. The MSIEI exhibited a decline followed by recovery, with moderate values expanding in the northern region in 2010. CONTAG generally increased, with moderate and high values becoming dominant by 2020, replacing low values in agro-forestry ecotones.

Consequently, the structural security map identifies central agricultural and urbanized areas as hotspots for habitat fragmentation risk (characterized by low to moderate security). Conversely, the northern and eastern forests represent high-security zones with minimal fragmentation. An overall improvement in structural security was observed in 2010, characterized by the expansion of moderately secure areas.

3.3.2. Process Security and Soil Erosion Risk

As shown in Figure 7, we analyzed the dynamics of process ecological security (characterized by soil erosion risk) using normalized LPI, NP, and MAI indices. Following normalization, higher values for all indicators indicate a higher level of process ecological security and a correspondingly lower risk of soil erosion.

Figure 7. Spatial patterns of process security characterized by soil erosion risks.

Spatially, high LPI values were continuous, while low values were distributed across the central and southeastern areas. The NP metric revealed a progressive increase in fragmentation: dominated by high values in 2000, the coverage of high and very high NP zones expanded to encompass almost the entire province by 2020, leaving few areas with moderate levels. MAI values were predominantly high, with localized clusters of low-to-medium values.

Integrating these metrics, the process security assessment reveals that the year 2000 represented the peak of ecological security, characterized by the most intact spatial structure and the strongest capacity to mitigate soil erosion. Although process security declined in 2010 and rebounded in 2020, it remained below the 2000 baseline. This trend reflects a landscape that, despite evident disturbances, retained a degree of resilience in mitigating soil erosion risks.

3.3.3. Function Security and Ecosystem Services Degradation Risk

As shown in Figure 8, we analyzed the dynamics of functional ecological security (characterized by the risk of ecosystem service degradation) using normalized LSI, SHDI, and AI indices. Following normalization, higher values for all indicators indicate a higher level of functional ecological security and a correspondingly lower risk of ecosystem service degradation.

Figure 8. Spatial patterns of functional security characterized by ecosystem service degradation risks.

Spatially, LSI values were predominantly high to very high, maintaining a stable spatial configuration that suggests consistent landscape shape complexity. The SHDI exhibited temporal fluctuations; initially dominated by medium-to-high values in 2000, it experienced a significant expansion of low-value zones in 2010 before recovering in 2020. The AI remained consistently high, reflecting a stable and highly aggregated landscape structure.

In terms of functional security, high-value zones formed a continuous expanse, contrasting with the scattered distribution of low-security areas. The temporal analysis shows that functional security was highest in 2000. Despite a significant decline in 2010 and a subsequent rebound in 2020, ecosystem services have not yet fully recovered to the 2000 baseline. This pattern illustrates a degradation-adjustment-recovery dynamic in the region’s ecological functions.

3.4. Reliability Analysis and Model Verification of Ecological Security Pattern Assessment Results

To validate the assessment results, Spearman correlation analysis was employed to examine the relationship between the ecological security pattern index (ESPI) and the comprehensive ecological quality (CHEQ) for the years 2000, 2010, and 2020 (Figure 9). Across all three time periods, a moderately positive and statistically significant correlation (p < 0.01) was observed between the ESPI and CHEQ, thereby confirming the reliability of the assessment findings.

Figure 9. The results of correlation analysis between ESPI and CHEQ. (The red dashed line indicates the positive correlation trend).

Furthermore, this study extended the validation to encompass assessment results across the three dimensions: structure, process, and function (Table 5). All correlation analyses yielded highly significant results (p < 0.01), providing additional evidence for the robustness of the assessment framework.

Table 5. Validation results of correlation analysis for ecological security pattern assessment.

Dimension	Variable Relationship	Year	Correlation Coefficient	Significance
	Richness—Habitat Fragmentation	2000	0.300	p < 0.01
Structure		2010	0.302	p < 0.01
		2020	0.283	p < 0.01
	Soil Erosion—Soil Conservation	2000	0.192	p < 0.01
Process		2010	0.178	p < 0.01
		2020	0.143	p < 0.01
	NPP—Functional Degradation	2000	0.407	p < 0.01
Function		2010	0.317	p < 0.01
		2020	0.298	p < 0.01

4. Discussion

4.1. Ecological Security Pattern Assessment Result Verification and Comparison

Validating the accuracy and reliability of assessment results is a critical step in determining the rationality and scientific rigor of an evaluation system [2]. However, most existing studies on ecological security assessment have neglected this validation phase [64]. Therefore, this study incorporated multiple independent external datasets to verify the credibility of the assessment results through correlation analysis. All associations were statistically significant (p < 0.01), although the correlation coefficients remained below 0.5. In large-scale spatial analyses, correlation values are often constrained by spatial heterogeneity and complex interacting factors. A meta-analysis by Møller and Jennions [65] supports this context. They noted that the mean effect size in ecological research is typically around r ≈ 0.19 due to the inherent noise in natural systems. Consequently, the coefficients observed in this study (0.143–0.435) indicate a statistically meaningful positive correspondence between the ESPI and the independent validation data. This confirms the reliability of the proposed framework.

The analysis reveals that the spatial pattern of ecological security in Heilongjiang Province derived in this study maintains a high degree of consistency with findings from prior research. For instance, Zhang et al. [66] observed significant spatial differentiation in the net ecosystem productivity (NEP) of Heilongjiang from 2010 to 2020, characterized by lower values in the eastern and western regions and higher values in the central and northern areas; this distribution aligns closely with our assessment results. Similarly, Sui et al. [16] identified that low-risk zones were predominantly concentrated in the northeast forest belt and the inner mongolia grassland regions. The spatial distribution of high-security and low-security areas identified in this study corresponds strongly with these findings, with high-security zones consistently clustered in the forested mountainous regions of the north and southeast. This spatial pattern is primarily attributed to the unique topography of the Greater and Lesser Khingan Mountains. As designated national key ecological function zones, these regions have benefited from the implementation of rigorous protection and restoration measures, notably the Plan for Ecological Protection and Economic Transformation of the Greater and Lesser Khingan Mountains Forest Region (2021–2035). These initiatives have significantly enhanced regional ecological quality and functions, thereby maintaining high vegetation coverage and ecosystem integrity, which in turn consolidates a stable ecological security barrier. Furthermore, Wu et al. [67] highlighted that high-risk areas in Heilongjiang were primarily concentrated in the Daqing region. Consistent with this, our study also characterizes Daqing City as an area with a notably low level of ecological security. As a major petroleum-resource-based city, Daqing faces severe challenges driven by large-scale oil extraction. These activities have caused widespread soil pollution, posing significant threats to human existence and ecosystem health [68]. This creates a sharp conflict between the imperatives of economic growth and the maintenance of ecological stability. Collectively, these three independent studies provide compelling evidence that the assessment framework constructed in this research possesses high reliability and that the adopted methodology is scientifically sound. Consequently, the assessment outcomes accurately reflect the actual status and spatial heterogeneity of ecological security in Heilongjiang Province.

In the future, the ecological security of Heilongjiang Province is expected to stabilize and improve. This positive trend is driven by national strategies, such as the Ecological Civilization initiative. However, global warming introduces new uncertainties. Specifically, the increasing frequency of droughts poses a significant risk to crop yields [69]. Therefore, it is essential to strengthen disaster monitoring and early warning systems. These measures are critical to mitigate emerging threats.

4.2. Assessment Framework for Ecological Security Pattern: Structure–Process–Function Integrating Risksz

The spatial configuration of a landscape is the product of intricate interplay between physical, biological, and anthropogenic forces. Turner [70] established that since human land use heavily imprints on landscapes, meaningful quantification of structure is a prerequisite to understanding pattern–process relationships. Wu [71] further identified the landscape as the nexus of human–nature interaction, essential for linking local phenomena to global scales. Based on these theories, we argue that landscape patterns determine regional ecological security by regulating ecological processes and functions. This lays the theoretical foundation for our proposed “structure–process–function” framework.

Many existing studies primarily rely on PSR models or DPSIR frameworks to assess ecological security. However, these approaches often lack geographical adaptability. They frequently overlook the specific disturbances and risks faced by a region [46,72]. Accordingly, we developed a structure–process–function framework that integrates ecological risks. We tailored this framework to the unique ecological assets and strategic needs of the study area, ensuring that the evaluation indicators are highly targeted. Furthermore, the framework couples specific risks with the three dimensions. Specifically, habitat fragmentation serves as a direct proxy for structural risk. It indicates the physical breakdown of ecological connectivity. Simultaneously, we address process risk by linking land use configurations to soil erosion. Ultimately, ecosystem service degradation is evaluated as the functional risk, capturing the final consequence of these structural and process disruptions. By integrating landscape metrics including LPI, NP, LSI, SHDI, and AI, we successfully quantified the spatial heterogeneity of regional risks. It transforms abstract ecological risks into a tangible, diagnosable assessment system. Consequently, monitoring metric trajectories enables the identification of ecological security tipping points. This provides a robust scientific basis for effective spatial governance.

This study is subject to certain limitations. Although we cross-validated our results using independent multi-source datasets, the lack of extensive ground observation data remains a constraint given the macroscopic scale of the study region. Moreover, we propose two directions for future research. On one hand, future studies could employ multi-scale analyses. This would help determine the optimal operational scales for different risk types. On the other hand, researchers could integrate this framework with climate change scenarios. This approach would enable the prediction of the future evolution of ecological security patterns.

5. Conclusions

Targeting the specific ecological context of Heilongjiang Province, this study constructed an assessment model integrating local risks within the structure–process–function framework. Cross-validation with independent data (NPP, soil conservation, habitat quality, and CHEQ) confirmed the model’s reliability. Subsequently, we investigated the spatiotemporal evolution of the ecological security pattern index (ESPI) from 2000 to 2020. The key findings are as follows:

(1)

The study revealed a “V-shaped” evolutionary trajectory of ecological security in Heilongjiang Province from 2000 to 2020, characterized by initial degradation followed by recovery. This trend validates the effectiveness of recent ecological restoration policies, indicating that macro-policy intervention has successfully mitigated the deterioration of the ecological environment.

(2)

The assessment highlighted that the mountainous regions in the north and southeast (Greater and Lesser Khingan Mountains) function as robust ecological barriers. In contrast, the plains and resource-dependent cities face security challenges. The distribution pattern indicates that the topographical factors and the transition of resource exploitation are the main factors leading to the regional security pattern.

(3)

By transforming ecological risks into diagnosable dimensions, this study demonstrates that ecological security is a dynamic coupling of physical structure, stable processes, and service functions. This study provides a scientific basis for differentiating spatial governance strategies.