The land use data in this article is derived from the national land use type remote sensing monitoring spatial distribution data provided by the Resource and Environment Science and Data Center of the Chinese Academy of Sciences (http://www.resdc.cn) (accessed on 3 February 2024), with a spatial resolution of 30 m × 30 m. The unified spatial reference coordinate system is the Krasovsky_1940_Albers coordinate system, including land use data for five periods: 1980, 1990, 2000, 2010, and 2020. The comprehensive accuracy of land use type evaluation in this dataset exceeds 90%.

Based on the “Current land use classification” (GB/T 21,010–2017) and considering the landscape characteristics of Heihe City, the landscape types in Heihe City are classified into seven categories: arable land, forest land, grassland, water bodies, urban and rural residential land, industrial and mining land, and unused land. Urban and rural residential land includes commercial and service land, industrial and mining storage land, residential land, public management and public service land, special land, and transportation and transportation land. Considering the spatial resolution of the image data and the precision of information extraction, other construction land in urban and rural areas, industrial and mining land, and residential land, including factory areas, large industrial areas, quarries, etc., are uniformly classified as industrial and mining land.

Landscape pattern indices can highly condense landscape spatial pattern information, reflecting its structural composition and spatial configuration characteristics (Han et al., 2018; Jingwei and Haize, 2006; Lu et al., 2012). Based on principles of landscape ecology, nine landscape indices were selected for calculation: at the patch scale level, the indices include LPI, AREA_MN, FRAC_AM, and COHESION; at the landscape scale, the indices include N.P., CONTAG, LPI, FRAC_AM, and SHDI (Table 1).

(1) Selection of Landscape Ecological Security Index

The Landscape Ecological Security Index construction method based on landscape disturbance and vulnerability has been widely applied in many studies (Feng et al., 2016; Nitschke and Innes, 2008; Sheng et al., 2018; Sun and Chen, 2017; Zhang et al., 2019). Landscape disturbance represents the degree of disturbance to different landscape ecosystems (Zhang et al., 2019), calculated by weighting landscape fragmentation, isolation, and dominance (Zhang et al., 2019). Landscape fragmentation describes the continuity and integrity of natural or anthropogenic landscapes, typically referring to the degree of fragmentation or isolation between different landscape elements (such as forests, grasslands, water bodies, etc.) within an area (Qing et al., 2007). Landscape isolation refers to the spatial distance and degree of isolation between different landscape types within a region (Jin and Hu, 2003). Landscape dominance is usually determined by comparing the coverage area of different landscape types (Cheng et al., 2005; Daqing et al., 2005; Jiang et al., 2002; Lohbeck et al., 2016). Existing research considers landscape fragmentation as the most important factor in landscape disturbance, followed by isolation and dominance, thus assigning values of 0.5, 0.3, and 0.2 (Gao et al., 2021; Wang et al., 2021), rrespectively. Landscape vulnerability represents the degree of change in landscape use types after disturbance (Wang et al., 2021). Based on existing research (Wang et al., 2021) and the actual situation of the study area, values were assigned as follows: urban and rural residential land is the most stable, assigned a value of 1; forests and grasslands are mostly located far from residential areas and relatively stable, assigned values of two and 3, respectively; farmland is susceptible to human disturbance, assigned a value of 4; water bodies, located close to residential and farmland areas, are greatly disturbed by human activities, assigned a value of 5; unused land and industrial and mining areas represent the most fragile ecosystems, assigned a value of 6. Using ArcGIS 10.3, the study area was divided into 1 km × 1 km grids, and each grid’s Landscape Ecological Security Index was calculated. Finally, the Landscape Ecological Security Index for the study area was obtained through Kriging interpolation (Table 2).

(2) Determination of Landscape Ecological Security Standards.

To reflect the spatiotemporal differences in the landscape ecological security index of the study area, reference to existing research results and combined with the actual distribution of the landscape ecological security index in the study area (Gao et al., 2021; Wang et al., 2021), the natural breakpoint method is adopted to divide the landscape ecological security level of the study area into five grades. The classification intervals for other years are standardized based on the classification intervals for the reference year (Table 3). To make the landscape ecological security pattern map more intuitive, each level of landscape ecological security is further divided into three equal parts while maintaining the same level intervals.

Global spatial autocorrelation can reveal whether geographic phenomena exhibit spatial clustering and is often characterized using Moran’s I (HuaLin, 2008). Local spatial autocorrelation reflects the degree of correlation between a geographic phenomenon or attribute in a local area unit within a larger region and the same phenomenon or attribute in adjacent local area units, and is often represented using LISA_i indices (HuaLin, 2008). The calculation formulas are as follows (Equations 1, 2).

(1) Global Spatial Autocorrelation:

In the equation, n represents the total number of sample points for variable x; x _i and x _j denote the observed values of a certain attribute in areas i and j, respectively; x ⃐ denotes the average value of the attribute across all n areas; W _ij represents the spatial weight matrix. When the Moran’s I index is positive, it indicates spatial clustering, meaning similar values are adjacent to each other; when the Moran’s I index is negative, it indicates spatial dispersion, meaning similar values are far apart from each other; when the Moran’s I index is close to 0, it indicates a random spatial distribution of patterns.

(2) Local Spatial Autocorrelation:

The LISA_i does not have a fixed range of values. A LISA_i value greater than 0 indicates spatial clustering of similar values around the area unit, manifested as high-high or low-low clustering. A LISA_i value less than 0 indicates spatial clustering of dissimilar values, manifested as high-low or low-high clustering.

Between 1980 and 2020, forest land, cultivated land, grassland, and unused land were the primary land-use types in Heihe City. Among them, forest land accounted for the largest area, though it steadily decreased over time, while the area of cultivated land and industrial mining land continuously expanded. The total area of industrial/mining land increased 32-fold, from 180.27 hm² in 1980 to 5,775.10 hm² in 2020. Spatially, forest land was widely distributed across various parts of Heihe City, with southern forest areas gradually being converted into farmland. Industrial and mining land was mainly concentrated in the southern and southwestern regions. Water bodies were primarily located in the southern part of the study area, while cultivated land was mainly distributed in the southwest and northeast. Unused land was more scattered, primarily dispersed in the southwestern part of the study area (Table 4; Figure 2).

Between 1980 and 2020, the mutual conversion among cultivated land, forest land, grassland, and unused land was the main direction of land use type transitions in the study area. During these 40 years, 3,924.13 hm² of forest land were converted to cultivated land, primarily in the southern and northeastern parts of the study area. Additionally, 2,064.5 hm² of grassland were converted to cultivated land, distributed throughout various parts of Heihe City. This indicates that, alongside mining development, Heihe City has increasingly emphasized the sustainability of agriculture and the ecological environment. Faced with the driving force of urban construction and development, 154.09 hm² of cultivated land were converted to urban and rural residential construction land, along with 84.45 hm² of forest land and 55.04 hm² of grassland, with the main conversion areas located in the southwestern part of Heihe City. It was observed that the degree of land use transfer for industrial and mining land was relatively low, with only a small portion converted to cultivated land. The main reason for this is Heihe City’s strong reliance on mineral resources, and the long-term environmental impacts caused by mining activities, which make it difficult for industrial and mining land to be repurposed for other uses in the short term after mining operations have ceased (Supplementary Table S1; Figure 3).

From the perspective of different land use periods, the main land use transitions occurred between 2000–2010 and 2010–2020. During 2000–2010, driven by the national policies of returning farmland to forest and grassland, along with the development of Heihe City’s forestry industry, the primary land transitions involved forest land. A total of 2,560.98 hm² of forest land was converted to cultivated land, 2,534.68 hm² to grassland, and 1,192.94 hm² to unused land. During the same period, 1,156.12 hm² of cultivated land were converted to forest land, while grassland and unused land contributed 2,444.57 hm² and 1,465.97 hm², respectively, to forest land (Supplementary Table S4; Figure 3). In the period between 2010 and 2020, the most significant land transitions involved water bodies, with 1,124.71 hm² of water area being converted to other land types, while only 573.73 hm² of other land were converted to water bodies. During this period, the water area was reduced by half, primarily transitioning into cultivated land and unused land. This reduction suggests potential over-extraction and processing of water resources, leading to their depletion and conversion into cultivated land. Additionally, the failure of ecological restoration attempts in areas affected by mining development may have resulted in the transformation of water bodies into unused land (Supplementary Table S5; Figure 3).

Overall, from 1980 to 2020, the changes in cultivated and forest land were the most significant, with a scattered spatial distribution. The general trend shows forest land undergoing the most notable transformations. The land use pattern exhibited significant changes in the northeastern and southwestern regions, while the northwestern and southeastern areas remained relatively stable (Supplementary Table S1; Figure 3).

Regarding patch contagion (COHESION), the patch indices for various land-use types were all close to 100, indicating a high degree of connectivity among the landscape patches, forming a relatively continuous and cohesive whole. However, the patch cohesion for urban and rural construction land and industrial and mining land was relatively low, though it gradually increased from 1980 to 2020. The primary reason for this increase is that Heihe City has vigorously developed its mining industry from 1980 to 2020, attracting a large number of mining-related industries and services, which led to the concentration of economic activities and, consequently, the focused distribution of industrial and mining land and urban-rural construction land. Additionally, the Heihe City government’s “General Land Use Plan for Heihe City (2006–2020)”(https://www.heihe.gov.cn/hhs/c102620/202012/c11_225272.shtml) (accessed on 8 October 2024) mandated adjustments in the scale of construction land, further enhancing the cohesion of urban-rural construction land and industrial and mining land (Figure 4).

In terms of area-weighted average patch fractal dimension (FRAC_AM), forest land and unused land exhibited relatively high values, close to 1.3. In contrast, the area-weighted average patch fractal dimensions for other land-use types were more balanced, hovering around 1.2 and 1.1. This indicates that the shapes of patches for forest land and unused land are relatively complex with jagged edges, while the patches for other land types have simpler shapes with smoother edges. By 2020, the area-weighted average patch fractal dimension for industrial and mining land had increased, mainly influenced by Heihe City’s ecological restoration policies. Human interference has caused changes in the shapes of these industrial and mining areas. The “General Planning for Mineral Resources in Heilongjiang Province (2016–2020)” (https://www.hlj.gov.cn/hlj/c108411/202110/c00_31181663.shtml) (accessed on 8 October 2024) proposed land reclamation in mining areas, which effectively addresses ecological issues in these regions, resulting in more complex shapes for industrial/mining land. Furthermore, the rugged terrain of Heihe City, characterized by rolling mountains and intersecting valleys, has contributed to the increasing complexity of the boundaries of industrial and mining land as mineral extraction deepens, which is another reason for the rise in the fractal dimension of these patches (Figure 4).

In terms of the largest patch index (LPI), forest land has consistently been the dominant landscape type in Heihe City. However, from 1990 to 2020, the LPI area of forest land gradually decreased, while the LPI area of cultivated land gradually increased. This shift can be attributed to Heihe City being primarily a forestry-based economy during this period, where economic development was closely tied to forest resources. As urban development progressed, activities such as deforestation and mining caused the originally continuous forest land to be fragmented into smaller patches of residential land, industrial and mining land, and unused land, resulting in spatial fragmentation (Figure 4).

Regarding average patch area (AREA_MN), forest land had the largest AREA_MN, indicating the lowest degree of landscape fragmentation. This is because forest land accounted for 60% of the original land-use types in Heihe City, and resource extraction in mining areas was concentrated in regions planned by the Heihe City government. As a result, forest areas that were either undeveloped or minimally impacted by development were often retained on the periphery of mining zones or in difficult-to-develop areas, leading to relatively concentrated patches of forest land. The AREA_MN of forest land peaked in 1990 and subsequently began to decline, mirroring the trend of increasing fragmentation in unused land and cultivated land during the same period. This indicates that after 1990, the process of urban integration contributed to the increased fragmentation of forest land, cultivated land, and unused land.

In terms of the area-weighted average patch fractal dimension (FRAC_AM), the study area showed a gradual decline from 1980 to 2020, decreasing by 0.02 over this period, indicating that the landscape in the study area has become simpler and the edges have become smoother. Regarding the largest patch index (LPI), the area of dominant landscapes in the study area gradually decreased from 1990 to 2010, but began to increase again from 2010 to 2020. This suggests a decline in the status of forest land as a dominant landscape, although it has somewhat recovered due to support from the return-to-forest policy. Concerning the number of patches (NP), there was a decline in the number of patches from 1980 to 1990, reaching a minimum value, indicating an increase in landscape connectivity in Heihe City in 1990, which enhanced the overall resilience of the landscape to disturbances. After 1990, the number of patches began to rise gradually, suggesting an increasing degree of landscape fragmentation in Heihe City. In terms of landscape diversity (SHDI), there has been an upward trend in landscape diversity in the study area since 1990. This increase is primarily associated with the rise in grassland area during the ecological restoration process, as well as the expansion of urbanization leading to an increase in the area of built-up and unused land. Regarding landscape contagion (CONTAG), the contagion index in the study area began to decline from 1990, dropping by 4% by 2010. This indicates that during this period, the degree of adjacency or connectivity between different patch types in the landscape of Heihe City weakened, while the extent of human activities affecting the natural landscape increased, coinciding with a gradual rise in urbanization levels.

From a general perspective, between 1980 and 2020, the spatial distribution of landscape ecological security levels in the study area showed a trend of being lower in the south and higher in the north. Over time, the overall landscape ecological security pattern has shown improvement. Specifically, in 1980, the proportions of low-security and relatively low-security areas in the study area were 28.37% and 36.03% respectively, mainly concentrated in the southern part with a fragmented landscape of agricultural land, forests, urban and rural construction land, and unused land. Areas classified as relatively safe and safe were primarily located in the north and east, covering 12.08% of the area, predominantly composed of forests with more intact land use. Overall, the ecological security level of Heihe City was relatively low in 1980. By 2020, the proportions of low-security and relatively low-security areas had decreased to 22.14% and 26.48% respectively, showing increases of 6.23% and 9.55%. This improvement was particularly notable in areas such as Aihui District, Wudalianchi City, and Bei’an City, likely due to integrated planning of agricultural land and controlled zoning for mineral development (Figure 6).

From 1980 to 1990, the ecological security levels in the study area continued to exhibit a pattern of lower levels in the south and higher levels in the north, but the ecological security in the low-security and relatively low-security areas in the southern and central regions improved. The proportion of low-security and relatively low-security areas decreased from 64.40% in 1980 to 41.11% in 1990, indicating an improvement in the ecological environment of Heihe City during this period, with an overall increase in landscape ecological security. From 1990 to 2000, the ecological security levels in the study area gradually shifted from the pattern of lower in the south and higher in the north to one of lower in the central region and higher in the surrounding areas. The environmental quality of Sunwu County improved, while that of Wudalianchi City and Nenjiang County deteriorated. From 2000 to 2010, the ecological security levels in the study area saw an overall improvement, forming a landscape ecological pattern of lower in the central region and higher in the surrounding areas, with particularly improved environmental quality in Wudalianchi City, where most areas were classified as relatively safe or safe. However, from 2010 to 2020, with the accelerated urbanization process, the ecological quality of Heihe City declined, and the overall ecological security level decreased. This decline was especially prominent in the central and southern regions, where areas classified as safe transitioned to relatively low-security areas, primarily due to an increase in unused land and a reduction in landscape restoration capacity (Figure 6).

From 1980 to 2020, the global spatial autocorrelation indices for the five periods were 0.892, 0.992, 1.029, 1.038, and 1.103, all of which passed the significance test at p = 0.01. This indicates that the ecological security index exhibits spatial clustering. Spatially, two extremes in Heihe City’s ecological security index were observed, with relatively few areas showing high-low or low-high clustering. Most areas exhibited no significant correlation, with local spatial autocorrelation primarily characterized by high-high value clustering and low-low value clustering.

In 1980, the clusters of ecological security indices were relatively scattered, with low-low value clusters appearing around the periphery. By 1990 and 2000, the spatial distribution of ecological security in Heihe City became more stable. Locally, high-high value clusters were mainly concentrated in the northern and eastern regions, indicating a relatively good level of ecological security in these areas. Meanwhile, the low-low value clusters around the edges gradually expanded southward, reflecting a decline in ecological security in the southern region, with a growing risk of ecological degradation. The spatial autocorrelation results for the ecological security index suggest that the northern and eastern regions, where forests are distributed, acted as ecological barriers, effectively maintaining the environment. In contrast, the central and southern regions, due to the development of industrial, mining, and urban construction land, experienced more frequent human disturbances, making the ecosystems more fragile. During the 2010 and 2020 periods, the area of high-high value clusters in the northern region became more stable, while the low-low value clusters in the central region shifted entirely to the south. This was due to the central region’s unused land patches gradually forming larger, more regular patches under proper urban planning, reducing ecological damage in the surrounding areas. However, urban and rural construction in the southern region began to impact the environment, highlighting the need for future attention to ecosystem restoration and protection in this area (Figure 7).

Evaluating the spatial pattern changes of mining cities using landscape pattern indices is meaningful and necessary because it can provide a reference for policymakers in planning mining cities (Wei et al., 2009). In selecting landscape pattern indices, covering every aspect in practical applications is optional. Landscape pattern indices serve repetitive purposes, and only a few indices are needed to fully meet the requirements of pattern analysis, reflecting the characteristics of landscape patterns in the study area. Therefore, this study screened landscape factors, and nine indices representing landscape fragmentation, aggregation, and diversity were selected for calculation.

Compared to the study by Ya et al., which only analyzed the spatial changes of the landscape from a temporal perspective (Ya-Hui et al., 2021), this study further divided landscape indices into patch-level and landscape-level analyses. It described the changes in landscape patterns of different land types in the study area over different time dimensions and the overall landscape pattern changes in the study area over different time dimensions, providing a comprehensive analysis from macro and micro perspectives, making the results more scientifically sound. From the results of landscape pattern analysis, it can be observed that human disturbances significantly impact the spatial pattern of a mining city.

Based on the analysis of landscape pattern indices, the landscape pattern of Heihe City has been primarily influenced by economic dependence, policy restrictions, and environmental issues.

From the patch scale perspective, significant changes were observed in the average patch area (AREA_MN) and the largest patch index (LPI) for forest land between 1980 and 2020. However, other landscape patterns showed little variation. This is mainly because the forestry industry holds an important position in the economic structure of Heihe City. The forestry industry is one of Heihe’s traditional sectors, with abundant forest resources, particularly high-quality timber such as red pine and white birch. The city has actively developed industries related to timber processing, non-timber forest product development, and forest tourism. With the growing awareness of sustainable development and ecological protection, Heihe City has gradually promoted the sustainable use of forest resources and the conservation of its forest resources in recent years. Over the past 40 years, mining activities and land use policies in Heihe City have been subject to strict government regulation and oversight. The local government has controlled the area of forest land, effectively increasing its extent by implementing afforestation projects on reclaimed farmland and unused land in various forest farms. Consequently, the changes in the landscape pattern of forest patches have been the most significant.

From the landscape scale perspective, the landscape diversity index of Heihe City showed an upward trend from 1980 to 2020, while other indices displayed a downward trend. This pattern is closely related to the mining policies, ecological restoration policies, and urbanization processes implemented in Heihe City. The city is rich in mineral resources, with a diverse range of types distributed across various patches, which facilitates economic diversification. Under this model of mining development, Heihe City has experienced significant changes in its economic structure, which has also influenced the updating and alteration of the landscape pattern. Additionally, due to the transformation of the industrial structure, the local government has adopted policies for the redevelopment and reuse of abandoned mining areas or extraction sites. This involves converting mined-out land into green spaces, parks, wetlands, or other forms of urban land. For instance, the Xigangzi coal mining area in Heihe City has combined mine environmental restoration with farmland occupation and compensation projects to transform industrial wasteland into fertile land. This has enriched the types of landscape patches within the city, thereby increasing landscape diversity. At the same time, Heihe City has initiated ecological restoration projects aimed at restoring ecosystems in mining areas through vegetation recovery, soil remediation, and water body management. These efforts have diversified the landscape structure of damaged areas and increased the complexity of the ecosystem. Furthermore, with the acceleration of the urbanization process, the government has expanded land types for various uses, such as commercial zones, residential areas, industrial districts, and transportation facilities. This urbanization process has clarified the boundaries between different land uses, increasing the number and variety of landscape patches and enhancing overall landscape diversity.

Due to the adoption of different standards, the results of landscape ecological security assessment of mining cities obtained from different research objects can only present conclusions partially consistent with those of previous studies (Chen and Ma, 2023). Scholars typically combined ecological security assessment with evaluation model studies in previous research. For instance, Yin et al. combined landscape ecological security assessment with the PSR model (Yin et al., 2012); While Zhang et al. integrated ecological security assessment with the Pressure-State-Response framework (Zhang R. et al., 2023). However, due to differences in evaluation objects and evaluation models, the scientific validity of the assessment results cannot be guaranteed. Additionally, evaluation criteria have become one of the controversial issues in landscape ecological security assessment results. For example, Wang et al. constructed an ecological security evaluation system through the evaluation of ecosystem service importance and ecological sensitivity, mainly using the Analytic Hierarchy Process (AHP) combined with expert questionnaires to obtain evaluation criteria, leading to highly subjective conclusions (Zhou et al., 2022). In this study, however, theoretical models were not relied upon, and all research was based on the land use changes in Heihe City from 1980 to 2020. In constructing ecological evaluation criteria, quantitative research methods were adopted, and the landscape ecological security assessment results obtained by calculating landscape pattern indices reduced the subjectivity and errors introduced by expert scoring methods.

Moreover, this study did not use a single ecological security evaluation system. Still, it supplemented the landscape ecological security assessment results through spatial autocorrelation analysis of landscape ecological security spaces, helping to deepen the understanding of the spatial characteristics and dynamic changes of ecosystems, thereby better assessing the health status of ecosystems and guiding ecological protection and management efforts. Spatial autocorrelation analysis of landscape ecological security spaces was used to examine the overall analysis of landscape ecological security indices, validating the feasibility of the ecological security assessment results.

Considering the economic development characteristics and natural environment conditions of Heihe City, and based on the 2020 ecological security evaluation results, corresponding management measures are proposed according to the different ecological security levels and characteristics, in order to reduce ecological security risks to a controllable range. The division of ecological functional areas from the perspective of sustainable development in mining cities has not yet been proposed for Heihe City (Shi and Xiang, 2023). This study, based on Heihe City’s ecological security index and higher-level planning, uses methods such as overlaying current land use data, DEM data, and administrative boundaries, along with spatial interpolation analysis, to divide Heihe City into four functional areas: the ecological security control area for mining cities, ecological protection area, ecological optimization area, and ecological comprehensive development area (Figure 8) (Song et al., 2023).

In the low and relatively low ecological security areas of Heihe City, there is a phenomenon of low-low value clustering, which is highly sensitive to human interference (Song et al., 2023). These clusters appear in agricultural and forested lands, where the ecological security status is poor, and serious ecological degradation and related issues are present. This region has been designated as the ecological security control area for mining cities. The total area of this control area is 4,308,956 hm², accounting for 64.52% of the total area. Key areas include the central and southern parts of Nenjiang County, Wudalianchi City, and Bei’an City, as well as the northern part of Nenjiang County and the central region of Aihui District. In this area, ecological regulation efforts should focus on reversing the trend of ecological retrogression, with particular attention to post-development land quality protection and soil restoration. Conservation and rehabilitation measures should be implemented to protect agricultural and forested lands. By constructing ecological corridors and restoring ecological nodes and barrier points, all development activities that could cause pollution should be prohibited, minimizing the impact of human activities on the existing ecological environment (Zhang et al., 2017; Zhang et al., 2018; Zhang et al., 2024; Li Q. et al., 2023).

The general ecological security area of Heihe City is in a state of ecological alert, where both low-high and high-low value clustering phenomena occur, indicating that the area is sensitive to human interference. Although the ecological environment has been partially damaged, it still retains some capacity for self-recovery (Feng et al., 2023). This region has been classified as the ecological security protection area for mining cities, mainly distributed in the northern parts of Xunke County, Sunwu County, and the northern part of Aihui District, covering an area of 15,948 hm². The dominant landscape types in this area include degraded unused land (bare rock), large expanses of high-vegetation-cover paddy fields, and human-disturbed forested areas (shrubland) and farmland. Therefore, priority should be given to protecting the degraded unused land and forested areas in this area, preventing the expansion of land use driven by the pursuit of economic gains, and reducing unreasonable human development activities. For the degraded unused land in this area, measures such as land leveling, increasing green cover, and prohibiting further development should be taken to maintain ecological security levels (Wei et al., 2023). As for the human-disturbed forested areas, strict implementation of the “Grain for Green” project should be enforced to prevent the over-expansion of farmland for economic development purposes.

The relatively safe areas of Heihe City are regions with strong resilience to disturbances, exhibiting high-high or high-low value clustering. These areas have an intact ecological structure and a high level of ecological security. This region has been designated as the ecological optimization area for mining cities, primarily located in Aihui District, covering a total area of 6,419 hm². The dominant landscape type in this area is undisturbed forestland, which can serve as an ecological barrier and support sustainable regional developments (Wei et al., 2023). Therefore, it is essential to continue promoting active forestland protection policies, implementing the “Grain for Green” and “Wetland Restoration” policies, and intensifying efforts to protect the forested landscape in this area (Feng et al., 2023). Human activity should be minimized to reduce negative impacts, and unreasonable agricultural practices such as deforestation for farming must be strictly limited. These measures will help maintain the natural ecological succession of Heihe City while balancing social, ecological, and economic benefits.

The safe areas in Heihe City are areas with strong self-recovery capabilities, where high-high value clustering is evident. These areas have the highest level of ecological security and can accommodate moderate development. This region has been designated as the ecological comprehensive development area, located in the northern part of Aihui District, covering a total area of 1,326 hm². Although this area enjoys a high level of ecological security, it also faces significant environmental pressures. In the course of development, efforts should leverage the concentration of forestland to enhance infrastructure development, promote the clustering of the forestry industry, and prioritize the advancement of modern forestry. Additionally, it is important to vigorously develop the service sector, accommodate industrial transfers from central cities, and manage population transfers from restricted and prohibited development areas. This approach aims to prevent ecological degradation caused by blind development.