A method for comprehensively identifying ecological sources of mining areas based on InVEST-MSPA double model fusion

Abstract

The application discloses a kind of based on InVEST-MSPA double model fusion's mine area ecological source place comprehensive identification method and system.Method is predicted by CA-Markov model multi-scenario land use pattern, identifies ecological hinge based on circuit theory, constructs ecological-mining game matrix and divides source type, and key node is screened by complex network node removal experiment.The method integrates dynamic scene and game analysis, breaks through the limitation of traditional static identification, improves the anti-disturbance ability of source and planning feasibility, adapts to the collaborative needs of mine area ecological protection and resource development.

Description

Technical Field

[0001] This invention belongs to the field of mining area ecology and relates to a comprehensive identification method for the source areas of mining area ecology based on the fusion of the InVEST-MSPA dual model. Background Technology

[0002] As one of the areas most severely affected by high-intensity human disturbance, the ecological protection and restoration of mining areas face unique challenges in terms of spatiotemporal dynamics. Unlike natural ecosystems, the land use patterns in mining areas exhibit significant dynamic evolution characteristics driven by mining activities. Mining sites advance year by year, spoil heaps dynamically accumulate, and reclaimed areas gradually recover. This complex process of "mining, destroying, and restoring simultaneously" keeps the structure and function of mining area ecosystems in constant flux. Currently, research on the identification of ecological source areas mainly employs static evaluation methods, such as ecosystem service function assessment based on the InVEST model and landscape pattern connectivity analysis based on the MSPA model. While these methods have achieved good results in conventional ecological planning, they have three fundamental limitations when applied to mining areas: First, there is a lack of dynamic consideration over time. Existing studies mostly identify source areas based on current land use data, ignoring the dynamic expansion trends of mining activities. An area with high ecological value at the present moment may be completely swallowed up by mining operations in the next few years; while an area that seems fragmented now may become a future ecological hub due to the reclamation of spoil heaps. This "status quo identification" model leads to a disconnect between planned ecological source areas and the mining process, making it difficult for them to survive sustainably amidst dynamic disturbances. Secondly, the interplay between ecology and mining is overlooked. Ecological protection and resource development constitute a core contradiction within mining areas. Existing research typically treats mining land simply as part of the ecological resistance surface, without deeply analyzing the interplay between ecological value and mining potential. Which areas should be permanently preserved? Which areas can be "mined first, then restored"? Which areas should be resolutely avoided? These questions lack scientific quantitative diagnostic tools. Third, connectivity analysis remains at the geometric morphology level. Traditional MSPA analysis, based on geometric morphology principles, divides the landscape into core areas, bridging areas, and other types. However, the movement of ecological flows in actual landscapes is not a simple geometric connection, but is influenced by a combination of various ecological resistances. Two patches that are geometrically connected may have their actual ecological flows completely interrupted if the area between them is filled with highly polluted or disturbed areas. This problem of "formally connected but actually disconnected" is particularly prominent in mining areas. In recent years, circuit theory has emerged as a promising approach in ecological network research. It abstracts the ecosystem as a conductive surface and identifies key channels and nodes of ecological flow by simulating the random walk paths of currents, providing a new perspective to overcome the limitations of traditional landscape geometric analysis. Meanwhile, the CA-Markov model is becoming increasingly mature in land use change simulation, effectively predicting future land use patterns. However, a mature solution for systematically integrating dynamic simulation, circuit theory, and game theory to construct an "anti-fragile" source area identification framework tailored to the specific characteristics of mining areas is currently lacking. Therefore, there is an urgent need for a dynamic identification method that can incorporate the "timeline" into source site identification, quantify the relationship between ecology and mining, and break through the limitations of geometric connectivity, in order to cope with the complex evolution of mining ecosystems under mining disturbances and truly identify those "antifragile" ecological source sites that can preserve and maintain connectivity during disturbances. Summary of the Invention

[0003] To address the problems existing in the background technology, this application proposes a method, system, equipment, and computer-readable storage medium for comprehensive identification of ecological source areas in mining areas based on the fusion of the InVEST-MSPA dual model. To achieve the above objectives, the technical solution adopted in this application is as follows: On the one hand, this invention provides a comprehensive identification method for the ecological source areas of mining areas based on the InVEST-MSPA dual-model fusion, including the following steps: S1. Obtain historical land use data, mining planning data and ecological protection data of the mining area, set up multi-scenario land use simulation schemes, and predict the land use pattern under different scenarios in future years based on the CA-Markov model. S2. Based on circuit theory, the ecosystem is constructed as a conductive surface model. The target species for ecological protection and the surrounding nature reserves are identified as voltage application points. The current density distribution is calculated, and key nodes where current convergence occurs are identified as ecological hubs. S3. Construct an ecological-mining game matrix, spatially overlay the ecological hubs identified in step S2 with the mining potential data, and classify different types of ecological source areas according to the combination relationship between ecological value level and mining potential level. S4. Construct a complex network model to simulate the impact of mining disturbances on the ecological network under different scenarios. Identify the key nodes that cause network collapse through node removal experiments and incorporate them into the final ecological source. Furthermore, the multi-scenario land use simulation scheme described in step S1 includes: Scenario A: Inertial Expansion Scenario, simulating continuous expansion of mining land according to historical expansion rates; Scenario B, a planning constraint scenario, involves strictly simulating mining operations according to the mining boundaries of the mining area; Scenario C, Ecological Priority Scenario, simulates a scenario where mining is prohibited within the ecological red line. Furthermore, in the conductive surface model described in step S2, areas with well-restored vegetation are set to low resistance values, while mining strippings and hardened road surfaces are set to high resistance values. Furthermore, the partitioning rules of the ecology-mining game matrix described in step S3 include: Areas with high ecological hub value and low mining potential are designated as permanent ecological source areas; Areas with high ecological hub value and high mining potential are analyzed using a time-for-space game based on the remaining years of mining. If mining ends within the preset time limit and there are conditions for rapid reclamation, they are designated as temporary source areas; otherwise, they are designated as areas to be avoided. Areas with low ecological value and low mining potential are designated as ecological reserve land. Furthermore, the preset time limit is 5 years. Furthermore, the node removal experiment in step S4 includes: simulating the change in connectivity of the ecological network after the disappearance of the core ecological source area, identifying those nodes that would cause the overall collapse of the network once removed, and determining them as key ecological source areas. Furthermore, for the aforementioned key ecological source areas, protective mining measures such as underground mining or backfilling mining are adopted. On the other hand, the present invention provides a comprehensive identification system for the ecological source areas of mining areas based on the InVEST-MSPA dual-model fusion, which operates the method described above, including: The data acquisition module is used to acquire historical land use data, mining planning data, and ecological protection data of the mining area; The scenario simulation module is connected to the data acquisition module and performs multi-scenario land use simulation based on the CA-Markov model to predict the land use pattern in future set years. The ecological hub identification module constructs a conductive surface model based on circuit theory, calculates the current density distribution, and identifies key nodes of the ecological hub. The game analysis module is connected to the ecological hub identification module to construct an ecological-mining game matrix and classify ecological source area types according to the combination relationship between ecological value and mining potential. The network resilience analysis module constructs complex network models, performs node removal experiments, and identifies key ecological sources that lead to network collapse. The source location determination module is connected to the game analysis module and the network resilience analysis module to comprehensively determine the final ecological source location of the mining area. Compared with the prior art, this application has the following beneficial effects: This approach shifts from static identification to dynamic game theory, introducing the time dimension into ecological source area identification. By using a CA-Markov model to simulate the expansion trend of mining land over the next 10-20 years, it predicts which ecological nodes will disappear as mining progresses. This future-reserved dynamic identification strategy allows planned ecological source areas to evolve in tandem with the mining process, avoiding the awkward situation of traditional static planning where source areas are "destroyed immediately after planning," and truly achieving spatiotemporal coordination between ecological protection and resource development. This innovative approach incorporates resource abundance assessment methods from economic geology, spatially overlaying ecological hub value with mining potential to classify different types of resource areas, including permanent source areas, temporary source areas, reserve land, and avoidance zones. This game theory diagnostic tool provides a scientific basis for delineating ecological protection red lines in mining areas, optimizing mining sequence, and planning reclamation, enabling decisions on "where to protect, where to mine, when to mine, and when to restore" to move from empirical judgment to quantitative analysis. This method replaces traditional MSPA geometric morphology analysis with circuit theory, abstracting the ecosystem as a conductive surface. By simulating the random walk path of current, it accurately identifies key convergence nodes of ecological flows. These nodes are "inescapable hubs" in species dispersal and ecological processes, playing a decisive role in maintaining the connectivity of the entire ecological network regardless of their geometric integrity. This method effectively solves the pseudo-connectivity problem of "visually connected but actually disconnected" in mining areas. By constructing a complex network model and conducting node removal experiments, the study simulates whether the ecological network will collapse after the disappearance of core source areas. Nodes that cause overall network instability upon removal are identified as key ecological source areas with "antifragile" properties. This innovative perspective not only focuses on the current ecological value of the source areas but also emphasizes their ability to maintain network integrity during disturbances, providing a new evaluation standard for the construction of ecological security patterns in mining areas. This framework systematically integrates CA-Markov dynamic simulation, circuit theory ecological flow analysis, game theory matrix conflict diagnosis, and complex network resilience assessment. The modules are logically progressive and mutually supportive, forming a complete "prediction-diagnosis-game theory-resilience" technology chain. This framework is not only applicable to the identification of ecological sources in mining areas but can also be extended to other regions affected by dynamic disturbances, such as urban fringe areas and areas affected by major engineering projects, demonstrating significant methodological value and broad prospects for application. For the identified critical ecological sources, especially those nodes whose removal would lead to network collapse, this invention explicitly proposes protective mining measures such as underground mining and backfilling mining. This closed-loop output model of "source identification - disturbance assessment - mining recommendation" enables the research results to directly serve mining process optimization and ecological compensation decisions, significantly enhancing the practical guiding value of the research. Through parallel analysis of multi-scenario simulations (inertial expansion, planning constraints, and ecological priority), this invention can assess the differences in the response of ecological networks under different mining strategies, providing a scientific basis for developing adaptive ecological restoration plans. Regardless of how the mining process evolves in the future, planners can quickly adjust ecological protection strategies based on the analysis results of this method to ensure the long-term stability of the ecological network. Attached Figure Description Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system structure diagram of the present invention. Detailed Implementation The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Example: To enable those skilled in the art to more clearly understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below using a large coal base in western my country as an example, combined with specific data. The total area of ​​this mining area is approximately 280 km², including multiple open-pit and underground mines in operation. Nature reserves and water conservation areas are distributed around it, making the conflict between mining activities and ecological protection particularly prominent. S1. Data Acquisition and Preprocessing: First, basic data collection was conducted. Remote sensing imagery was used to acquire four periods of Landsat-8 / 9 OLI imagery (30m resolution) and Sentinel-2 imagery (10m resolution) for the study area in 2010, 2015, 2020, and 2024, for land use classification and vegetation index extraction. Land use data, historical land use maps of the mining area, mining rights boundaries, mining site red lines, spoil heap areas, and reclamation area areas (from mining enterprises and natural resources departments). Ecological protection data, including the boundaries of nature reserves surrounding the mining area, the scope of water source protection areas, and ecological red line delineation plans (from the ecological and environmental protection department). Mining planning data, collection of mining plans for the next 10-15 years, mine succession arrangements, and spoil heap plans (from mining companies). Digital elevation model and ASTER GDEM 30m resolution data were used for terrain analysis. Data preprocessing was performed. ENVI software was used to perform radiometric calibration, atmospheric correction, and geometric fine correction on the remote sensing imagery. An object-oriented classification method was adopted, combined with field survey sampling points (300 sampling points), to generate four land use / cover classification maps for 2010-2024. The classification system includes eight categories: cultivated land, forest land, grassland, water area, construction land, mining sites, spoil heaps, and bare land, with an overall classification accuracy exceeding 92%. The Normalized Difference Vegetation Index (NDVI) was calculated to extract vegetation cover information. All data were unified to the CGCS2000_3_Degree_GK_Zone_36 projection coordinate system, and the raster resolution was resampled to 30m. S2, Multi-scenario Land Use Simulation (Introducing a Time Axis). CA-Markov model construction. Based on land use change data from 2010 to 2020, the land use transition probability matrix is ​​calculated. Driving factors such as elevation, slope, distance from mining site, distance from road, distance from settlement, distance from river, and mining rights boundary are selected to construct a suitability atlas. The CA-Markov model was run in the IDRISI software, with 2020 as the base year, to simulate the land use pattern in 2024, and the accuracy was verified by comparing it with the actual 2024 data (Kappa coefficient = 0.86, which meets the accuracy requirements). Configure multi-scenario solutions. Scenario A (Inertial Expansion): Assuming that mining land continues to expand outward at the average expansion rate from 2010 to 2020 (3.2 km² per year), without being restricted by planning boundaries, the simulation continues until 2035. Scenario B (Planning Constraints): Strictly adhering to the existing mining rights boundaries and mining plans of the mining area, the mining site will be advanced year by year according to the plan, and the spoil heap will be piled up according to the design, simulating up to 2035. Scenario C (Ecological Priority): Based on Scenario B, nature reserves, core areas of water source protection areas, and areas within ecological red lines are designated as no-mining zones, and the direction of mining land expansion is adjusted, simulating up to 2035. Land use pattern maps for 2035 under three scenarios were generated. Under Scenario A, mining land will expand to 85.6 km² (30.6%); under Scenario B, it will expand to 72.3 km² (25.8%); and under Scenario C, it will expand to 65.8 km² (23.5%), while ecological protection land will be effectively preserved. S3. Ecological hub identification based on circuit theory. Ecological resistance surface construction. The black stork (a large wading bird with a wide range of activities and high requirements for habitat connectivity), a typical protected species in the mining area, was selected as the target species. Based on relevant literature and expert consultation, resistance values ​​were set for different land use types: forest land 1, grassland 5, cultivated land 10, water area 20, bare land 50, construction land 200, mining area 500, and spoil heap 300 (the higher the resistance value, the greater the difficulty of ecological flow). Simultaneously, factors such as slope and distance from water source are considered to correct the resistance value, generating a comprehensive ecological resistance surface raster map. Voltage application point settings. Two nature reserves and one water source protection area around the mining area are set as voltage sources (1V), and the existing ecologically important areas inside the mining area (identified based on the InVEST model) are set as voltage grounding terminals (0V). Using Circuitscape software, the random travel paths of electrons on the resistive surface were simulated based on circuit theory. Running the "pairwise" mode, the current density distribution between all voltage sources and the ground terminal was calculated. A current density map was output; regions with high current density in the map represent key channels and convergence nodes of the ecological flow. The current density was normalized, and the top 10% of the grid cells with the highest current density values ​​were extracted. After converting the grid to vector and removing scattered patches with an area smaller than 0.5 km², 18 ecological hub patches were obtained, with a total area of ​​12.8 km². These hubs are mainly distributed in the southeastern river valley of the mining area, the northwestern natural secondary forest area, and key corridors connecting the protected area. S4. Construction of Eco-Mining Game Matrix and Conflict Diagnosis. Data on coal resource reserves distribution, coal seam depth, stripping ratio, and mineable thickness were collected from the mining area. Using a resource abundance evaluation method, the above indicators were normalized and weighted to generate a mining potential distribution map (values ​​range from 0 to 1, with higher values ​​representing greater mining value). Mining potential was divided into three levels: high potential (>0.7), medium potential (0.4-0.7), and low potential (<0.4). The ecological hub patches identified in S3 are designated as high ecological value areas (value 3). Other important ecological patches retained under scenario C in S2 (such as the core area and bridging area) are designated as medium ecological value areas (value 2). The remaining areas are designated as low ecological value areas (value 1). Spatially overlaying ecological value levels with mining potential levels generates a conflict diagnosis map. For five patches (total area 4.0 km²) with high ecological value and high mining potential, mining plans are retrieved: Patch E3 is located on the southeast side of an open-pit mine and is planned to be mined out within 5 years. After mining, internal spoil reclamation will be carried out, with a reclamation cycle of approximately 3 years. According to the assessment, the ecological function can be restored after 2030, and it is recommended that it be designated as a temporary source area, with mining and reclamation measures implemented during the mining period. Patch E7 is located above a certain well mine and is planned to be mined in 8 years. Using backfilling mining technology can significantly reduce surface subsidence. It is recommended to try to avoid it or adopt protective mining. S5. Complex Network Model and Node Removal Experiment (Resilience Analysis). The permanent source areas, temporary source areas, and ecological reserve land identified by S4 were used as nodes (a total of 23 nodes, with an area >1km²). The current density calculated by circuit theory was used as the connection strength, and grids with a current density >0.05 were extracted as potential corridors to construct the ecological network topology. A stepwise removal strategy is employed, removing one node at a time (simulating the disappearance of the source region due to mining), and calculating the overall connectivity index of the remaining network (such as network connectivity, average shortest path length, clustering coefficient, etc.). Special attention is paid to nodes whose removal results in a network connectivity decrease of more than 50% or the network splitting into multiple subnets. Experimental results show that once nodes P2 (located in the central river valley of the mining area, serving as a hub connecting the two major protected areas in the north and south), P5 (the core area of ​​the natural forest in the southeast, with the largest area), and P9 (the western water conservation area) are removed, the overall network connectivity decreases by 72%, 68%, and 61%, respectively, and the network splits into 3-4 isolated subnets. These three nodes have been identified as key "antifragile" ecological sources, and are irreplaceable. Node P5 is located above a certain underground mine. The coal seam is deep, and the impact of mining is relatively small. It is recommended to use strip mining or backfilling mining to ensure that the surface ecological function is not lost. Node P9 is located at the boundary of the mining area and has no mining plan; it can be directly included in the permanent protection zone. Node P2 is located at the edge of the planned mining area. It is recommended to adjust the mining sequence, reserve a sufficiently wide ecological corridor, or adopt underground mining methods. S6. Identification of the final source and recommendations for countermeasures. By merging the 7 permanent source areas (S4) and 3 key nodes (2 of which are already included in the permanent source areas, and 1 new P2 node), a total of 8 ecological source areas are obtained, covering a total area of ​​8.2 km². The 4 temporary source areas (S4, covering a total area of ​​3.2 km²) will be used as dynamically managed source areas, with phased protection and restoration plans developed. The S4 ecological reserve land (12.5 km²) will be included in the ecological restoration potential area. The following tiered control measures are implemented: Tier I Control Zone (Permanent Ecological Source Area): All mining activities are prohibited. A buffer zone is established around the zone, and ecological monitoring and restoration are carried out to ensure the continuous improvement of ecological functions. Tier II Control Zone (Temporary Source Area): Mining is permitted during the planning period, but protective mining measures (such as backfilling and height-restricted mining) must be implemented, and a detailed reclamation plan must be developed. Ecological reconstruction must be completed within two years after mining. Tier III Control Zone (Ecological Reserve Land): This can serve as a backup area for future mining, but the existing ecological functions must be maintained, and an ecological impact assessment must be conducted before mining. Multiple backup corridors are constructed between key ecological nodes, and alternative paths are identified using circuit theory to improve network redundancy. For the areas around node P2 that are affected by mining, implement ecological restoration projects in advance to enhance their buffering capacity during disturbances. The identification results of this method are compared with those of the traditional static MSPA+InVEST method: Traditional methods identified 15 ecological source areas, covering a total area of ​​12.3 km², of which 5 (40%) are located within planned mining areas for the next 10 years and do not have long-term sustainability. This method identified 8 permanent source sites, all of which avoided future mining planning areas; 4 temporary source sites, achieving time-series coordination between resource development and ecological protection; and the successful identification of the critical node P2 avoided network collapse that might have been caused by traditional geometric analysis neglecting its pivotal value. By introducing time axis, circuit theory and game theory matrix, this embodiment successfully achieves dynamic and accurate identification of ecological source areas in mining areas, providing a scientific basis and technical support for the coordinated planning of ecological protection and resource development in mining areas. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A comprehensive identification method for ecological source areas in mining areas based on the fusion of InVEST-MSPA dual models, characterized in that, Includes the following steps: S1. Obtain historical land use data, mining planning data and ecological protection data of the mining area, set up multi-scenario land use simulation schemes, and predict the land use pattern under different scenarios in future years based on the CA-Markov model. S2. Based on circuit theory, the ecosystem is constructed as a conductive surface model. The target species for ecological protection and the surrounding nature reserves are identified as voltage application points. The current density distribution is calculated, and key nodes where current convergence occurs are identified as ecological hubs. S3. Construct an ecological-mining game matrix, spatially overlay the ecological hubs identified in step S2 with the mining potential data, and classify different types of ecological source areas according to the combination relationship between ecological value level and mining potential level. S4. Construct a complex network model to simulate the impact of mining disturbances on the ecological network under different scenarios. Identify the key nodes that cause network collapse through node removal experiments and incorporate them into the final ecological source.

2. The method according to claim 1, characterized in that, The multi-scenario land use simulation scheme mentioned in step S1 includes: Scenario A: Inertial Expansion Scenario, simulating continuous expansion of mining land according to historical expansion rates; Scenario B, a planning constraint scenario, involves strictly simulating mining operations according to the mining boundaries of the mining area; Scenario C, Ecological Priority Scenario, simulates a scenario where mining is prohibited within the ecological red line.

3. The method according to claim 1, characterized in that, In the conductive surface model described in step S2, areas with well-restored vegetation are set to low resistance values, while mining strippings and hardened road surfaces are set to high resistance values.

4. The method according to claim 1, characterized in that, The partitioning rules of the ecology-mining game matrix mentioned in step S3 include: Areas with high ecological hub value and low mining potential are designated as permanent ecological source areas; Areas with high ecological hub value and high mining potential are analyzed using a time-for-space game based on the remaining years of mining. If mining ends within the preset time limit and there are conditions for rapid reclamation, they are designated as temporary source areas; otherwise, they are designated as areas to be avoided. Areas with low ecological value and low mining potential are designated as ecological reserve land.

5. The method according to claim 4, characterized in that, The preset time limit is 5 years.

6. The method according to claim 1, characterized in that, The node removal experiment described in step S4 includes: simulating the changes in the connectivity of the ecological network after the disappearance of the core ecological source area, identifying those nodes that would cause the entire network to collapse once removed, and determining them as key ecological source areas.

7. The method according to claim 6, characterized in that, For the aforementioned key ecological source areas, protective mining measures such as underground mining or backfilling mining are adopted.

8. A comprehensive identification system for the ecological source areas of mining areas based on the InVEST-MSPA dual-model fusion, characterized in that, include: The data acquisition module is used to acquire historical land use data, mining planning data, and ecological protection data of the mining area; The scenario simulation module is connected to the data acquisition module and performs multi-scenario land use simulation based on the CA-Markov model to predict the land use pattern in future set years. The ecological hub identification module constructs a conductive surface model based on circuit theory, calculates the current density distribution, and identifies key nodes of the ecological hub. The game analysis module is connected to the ecological hub identification module to construct an ecological-mining game matrix and classify ecological source area types according to the combination relationship between ecological value and mining potential. The network resilience analysis module constructs complex network models, performs node removal experiments, and identifies key ecological sources that lead to network collapse. The source location determination module is connected to the game analysis module and the network resilience analysis module to comprehensively determine the final ecological source location of the mining area.

Images