Method and device for generating restoration strategy of watershed landscape ecological protection area

Abstract

The invention discloses a watershed landscape ecological protection area restoration strategy generation method and device, and relates to the technical field of ecological optimization, and the method comprises the steps: predicting the multi-scene area demand data of a target watershed based on the actual area vector and transition probability matrix of each land utilization category extracted from a reference year land utilization grid map; based on the reference year land utilization grid map, the first driving factor set, the multi-scene area demand data, the neighborhood factor and the probability transfer matrix, predicting a future land utilization grid map through an FLUS model; constructing a multi-scene dynamic ecological security pattern based on a future land utilization grid map; and obtaining a driving factor contribution degree sorting table, a factor interaction type matrix and a key risk driving combination identification result based on the multi-scene dynamic ecological security pattern and the second driving factor set, and further determining a target ecological restoration protection area to generate a restoration strategy for the target ecological restoration protection area.

Description

Technical Field

[0001] This application relates to the field of ecological optimization technology, and in particular to a method and apparatus for generating restoration strategies for watershed landscape ecological protection zones. Background Technology

[0002] In related technologies, the spatial hierarchy for constructing ecological security patterns is mainly concentrated at the provincial, municipal, and county levels. The boundaries of the study areas are limited to specific administrative boundaries, lacking research on cross-scale and cross-administrative boundaries, such as river basins, urban agglomerations, and larger areas. Moreover, ecological elements outside administrative boundaries may still affect the ecological functions and processes within the region, thereby disturbing the ecological security pattern.

[0003] Furthermore, the selection of correction parameters is also a major challenge in constructing an ecological security pattern. Usually, based on land use and cover type, expert scoring is used to directly assign values. However, this method is highly subjective and cannot truly reflect the spatial differentiation of ecological resistance. It cannot accurately characterize the resistance that may be encountered in the ecological process. Moreover, in areas with large topographic relief, frequent geological disasters, and fragmented landscape patterns, the direct assignment method will ignore the impact of natural environmental factors on the ecological security pattern.

[0004] Moreover, the construction of ecological security patterns often takes a certain time as a node, lacking exploration of long-term series and different future scenarios. Research on the spatiotemporal evolution of regional ecological security patterns is still somewhat insufficient, making it impossible to accurately predict future land use grid maps, and consequently unable to generate watershed landscape ecological protection zone restoration strategies based on future land use grid maps. Summary of the Invention

[0005] In view of this, this application provides a method and apparatus for generating restoration strategies for watershed landscape ecological protection zones.

[0006] The objective of this application can be achieved through the following technical solutions: The first aspect of this application is to provide a method for generating restoration strategies for watershed landscape ecological protection zones, including: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps; Determine the transition probability matrix based on historical land use raster maps; Based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, multi-scenario area demand data for the target watershed are predicted. Based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, the future land use raster map is predicted by the FLUS model. The first set of driving factors has the same resolution and range as the base year land use raster map. Constructing a dynamic ecological security pattern based on future land use grid maps; Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, we obtained the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The second set of driving factors is the driving factor corresponding to the future land use raster map. The target ecological restoration protected area was determined based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations. Generate restoration strategies for the target ecological restoration protected area.

[0007] In one alternative embodiment, determining the migration probability matrix based on a historical land use raster map includes: For any two historical land use raster maps, count the number of pixels for the initial land use category and the number of pixels for the final land use category in the historical land use raster map; The transition probability is calculated based on the number of pixels in the initial land use category and the number of pixels in the final land use category. Generate the transition probability matrix based on the transition probability.

[0008] In one optional embodiment, a multi-scenario dynamic ecological security pattern is constructed based on a future land use raster map, including: Based on the future land use raster map and multi-source environmental data of the target watershed, ecological source areas are identified and a binary map of ecological source areas is obtained. The corrected ecological resistance coefficient raster map was determined based on the future land use raster map and multi-source environmental data of the target watershed. Ecological corridors and strategic nodes are identified based on the binary map of the target ecological source area and the raster map of the corrected ecological resistance coefficient.

[0009] In one optional embodiment, ecological source areas are identified based on a future land use raster map and multi-source environmental data of the target watershed, resulting in a binary map of the ecological source areas, including: The future land use raster map is converted into a binary map, which includes foreground and background areas; The core area is determined by performing morphological operations on the foreground region using the eight-neighbor algorithm, and the core area is identified as a candidate area for the ecological source area. Obtain a spatial distribution map of at least one ecosystem service indicator based on multi-source environmental data of the target watershed; The weights of each ecosystem service indicator were determined using the analytic hierarchy process (AHP), and the spatial distribution map was weighted and superimposed using these weights to generate an ecosystem importance score map. By spatially masking the comprehensive ecological importance score map with the core area, regions in the core area with ecological importance scores higher than the preset score threshold are selected, and a binary map of the target ecological source area is generated based on the region.

[0010] In one optional embodiment, determining a modified ecological resistance coefficient raster map based on a future land use raster map and multi-source environmental data of the target watershed includes: Based on future land use raster maps and multi-source environmental data of the target watershed, landscape structure index raster maps, landscape vulnerability index raster maps, and landscape loss index raster maps are generated. Based on the landscape structure index raster map, the landscape vulnerability index raster map, and the landscape loss index raster map, the landscape ecological risk index raster map is calculated. The basic ecological resistance coefficient for each land use type is determined based on the future land use grid map and local ecological planning standards. A basic ecological resistance raster map is generated based on the basic ecological resistance coefficient. Obtain at least one hazard-causing factor; Generate a geological hazard sensitivity raster map based on at least one hazard-causing factor; An average geological hazard sensitivity raster map is generated based on the geological hazard sensitivity raster map and the future land use raster map; Based on the basic ecological resistance raster map, landscape ecological risk index raster map, geological hazard sensitivity raster map, and average geological hazard sensitivity raster map, the corrected ecological resistance coefficient raster map is calculated.

[0011] In one optional embodiment, based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, a ranking table of driving factor contributions, a matrix of factor interaction types, and the identification results of key risk driving combinations are obtained, including: Based on the dynamic ecological security pattern of multiple scenarios, the dependent variable is determined, and the second driving factor is used as the independent variable. The geographic detector is used for detection and analysis to obtain the driving factor contribution ranking table, the factor interaction type matrix, and the key risk driving combination identification results.

[0012] In one optional embodiment, the target ecological restoration protected area is determined based on a ranking table of driving factor contributions, a matrix of factor interaction types, and the identification results of key risk-driven combinations, including: Target driving factors are determined based on the driving factor contribution ranking table. Obtain the spatial raster map corresponding to the target driving factor; Identifying target factor combinations based on factor interaction type matrix; A collaborative risk raster is obtained by overlaying the spatial raster maps corresponding to the target driving factors based on the combination of target factors. Target ecological restoration protected areas were determined based on the results of key risk-driven portfolio identification and collaborative risk raster map.

[0013] The second aspect of this application is to provide a device for generating restoration strategies for watershed landscape ecological protection zones, comprising: The acquisition module is used to acquire multi-period historical land use raster maps of the target watershed and determine the base year land use raster map from the historical land use raster maps; The first determining module is used to determine the transfer probability matrix based on the historical land use raster map; The first prediction module is used to predict multi-scenario area demand data for the target watershed based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map. The second prediction module is used to predict the future land use raster map based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, using the FLUS model. The first set of driving factors has the same resolution and range as the base year land use raster map. The module is used to construct a dynamic ecological security pattern for multiple scenarios based on future land use raster maps; The second determination module is used to obtain the driving factor contribution ranking table, factor interaction type matrix and key risk driving combination identification results based on the multi-scenario dynamic ecological security pattern and the second driving factor set. The second driving factor set is the driving factor corresponding to the future land use raster map. The third determination module is used to determine the target ecological restoration protected area based on the ranking table of driving factor contribution, the matrix of factor interaction types, and the identification results of key risk driving combination. The generation module is used to generate restoration strategies for the target ecological restoration protected area.

[0014] A third aspect of this application is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the method as described in the first aspect.

[0015] A fourth aspect of this application is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method as described in the first aspect.

[0016] Compared with existing technologies, the watershed landscape ecological protection zone restoration strategy generation method provided in this application predicts multi-scenario area demand data of the target watershed based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map; predicts the future land use raster map using the FLUS model based on the base year land use raster map, the first set of driving factors, the multi-scenario area demand data, neighborhood factors, and probability transition matrices; constructs a multi-scenario dynamic ecological security pattern based on the future land use raster map; and obtains a driving factor contribution ranking table, a factor interaction type matrix, and key risk driving combination identification results based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, thereby determining the target ecological restoration protection zone and generating restoration strategies for the target ecological restoration protection zone. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a method for generating restoration strategies for watershed landscape ecological protection zones provided in this application embodiment; Figure 2 A structural block diagram of the watershed landscape ecological protection zone restoration strategy generation device provided in this application embodiment; Figure 3 This is a structural block diagram of an electronic device for implementing a method for generating restoration strategies for watershed landscape ecological protection zones, as provided in an embodiment of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present 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 application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present application.

[0020] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] It should be understood that in the embodiments of this application, "at least one" means one or more, and "more than one" means two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the related objects before and after it are in an "or" relationship. "Contains A, B and / or C" means containing any one, two, or three of A, B, and C.

[0022] It should be understood that in the embodiments of this application, "B corresponding to A", "B corresponding to A", "A corresponds to B", or "B corresponds to A" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0023] To address the technical problems existing in related technologies, this application provides a method and apparatus for generating restoration strategies for watershed landscape ecological protection zones.

[0024] The method for generating restoration strategies for watershed landscape ecological protection zones provided in this application can be executed by an electronic device, such as a terminal or a server. The terminal can be a smartphone, tablet, laptop, or other similar device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. It is understood that this application does not limit the specific entity executing the method for generating restoration strategies for watershed landscape ecological protection zones.

[0025] The technical solution of this application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments described below are used to explain the technical solution of this application and are not intended to limit actual use.

[0026] To address the technical problems existing in related technologies, embodiments of this application provide a method for generating restoration strategies for watershed landscape ecological protection zones, such as... Figure 1 As shown, Figure 1 This is a flowchart illustrating a method for generating restoration strategies for watershed landscape ecological protection zones, as provided in an embodiment of this application. It should be noted that the steps shown may be executed in a different logical order than those depicted in the flowchart. The method may include the following steps S101 to S108.

[0027] Step S101: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps.

[0028] It should be noted that focusing on multi-scale, cross-administrative-boundary watersheds and deeply analyzing the coupling relationships between ecological processes is one of the key breakthrough directions for constructing an ecological security pattern. This approach is more conducive to solving multi-level ecological security problems and is an important way to coordinate the systematic governance of mountains, rivers, forests, fields, lakes, grasslands, and deserts. Multi-period historical land use raster maps refer to land use spatial data acquired in different years (e.g., 2000, 2010, and 2020) covering the same geographical area and using the same classification system. The format is raster, with each cell representing a land use category; for example, forest = 1, cultivated land = 2, construction land = 3, etc. The base year land use raster map usually refers to the latest period in the multi-period historical land use raster map, serving as the starting state for future land use simulations and the model training label.

[0029] Step S102: Determine the transition probability matrix based on the historical land use raster map.

[0030] In one optional embodiment, the transition probability matrix is ​​determined based on historical land use raster maps, specifically including the following steps: for any two periods of historical land use raster maps, the number of pixels in the initial land use category and the number of pixels in the final land use category of the historical land use raster maps are counted; the transition probability is calculated based on the number of pixels in the initial land use category and the number of pixels in the final land use category; and a transition probability matrix is ​​generated based on the transition probability.

[0031] It should be noted that any two historical land use raster maps have the same spatial extent and projection coordinate system, the same classification system, and the same resolution.

[0032] In one specific embodiment, for any two historical land use raster maps (e.g., 2010-2020), the two historical land use raster maps are matched one-to-one according to the pixel position, and the number of pixels of the initial land use category and the number of pixels of the ending land use category of the historical land use raster map are counted to obtain the transition frequency matrix; each row in the transition frequency matrix is ​​normalized to obtain the transition probability; and a transition probability matrix is ​​generated based on the transition probability.

[0033] The transition frequency matrix is ​​shown in the table below: Table 1

[0034] Step S103: Based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, predict the multi-scenario area demand data for the target watershed.

[0035] In an optional embodiment, the watershed landscape ecological protection zone restoration strategy generation method provided in this application further includes: adjusting the transition probability matrix according to the context.

[0036] In one specific embodiment, the scenario may include a natural growth scenario, an ecological protection scenario, and an urban development scenario. It should be noted that a natural growth scenario means not intentionally changing land use categories, allowing them to evolve naturally according to current conditions. An ecological protection scenario involves designating ecological protection zones, strengthening the protection of ecological land such as forests, grasslands, and water bodies, while reducing the expansion capacity of other land use categories. An urban development scenario involves improving the expansion and transfer capacity of arable land and construction land to meet the needs of urban development.

[0037] In one specific embodiment, under an ecological protection context, the following is set up: In the context of urban development, Set to original 1.5 times.

[0038] In one alternative embodiment, based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, a Markov model is used to predict multi-scenario area demand data for the target watershed: (1); in, This represents the predicted area of ​​each land use category after k years from the base year. P represents the area of ​​each land use category at base year t. This represents the transition probability matrix, indicating the probability of type conversion between various land use categories.

[0039] Step S104: Based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, predict the future land use raster map using the FLUS model.

[0040] It should be noted that the first set of driving factors has the same resolution and range as the land use raster map of the base year.

[0041] In one optional embodiment, the first set of driving factors includes at least one of the following: slope, aspect, rainfall, air temperature, land surface temperature, net primary productivity of vegetation, nighttime light index, soil type, and landform type. The driving factors are dynamically updated according to the scenario. That is, for each scenario, the driving factors are dynamically adjusted or regenerated to realistically reflect the future socio-economic, natural environmental, or policy constraints under that scenario, thereby ensuring that the simulation results are consistent with the scenario and realistically reasonable.

[0042] In another optional embodiment, the watershed landscape ecological protection zone restoration strategy generation method provided in this application embodiment further includes: determining the transition matrix based on the probability transition matrix.

[0043] It should be noted that the transition matrix is ​​represented by 0 and 1. When a land use type is allowed to be converted to another land use type, the corresponding value of the matrix is ​​set to 1, and when the conversion is not allowed, it is set to 0.

[0044] In one specific embodiment, if the probability in the probability transition matrix is ​​greater than a preset probability threshold, the value corresponding to the position of that probability is set to 1; otherwise, it is set to 0. The preset probability threshold is determined based on historical land use data and local policies.

[0045] In another alternative embodiment, the neighborhood factor is used to represent the ease with which a certain land use type can be converted to other land use types. The parameter ranges from 0 to 1, with a value closer to 1 indicating a stronger expansion capacity for that land use category. The neighborhood factor is determined based on historical land use data.

[0046] In another alternative embodiment, the FLUS (Flow-directed Landscape Unit Segmentation) model is a land use prediction model based on cellular automata. It can effectively simulate the spatial pattern of land use in different years and under different scenarios and spatially allocate the simulated land use quantity demand based on the transformation rules and quantitative relationships between land use categories and driving factors and the roulette wheel selection algorithm.

[0047] By inputting the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors, and transition matrix into the FLUS model, the future land use raster map is predicted.

[0048] In this step, the Markov model focuses on the time dimension to analyze land use change prediction models, which has the advantage of high efficiency in long-term quantitative prediction. When combined with the FLUS model, it can not only improve the prediction accuracy of land use quantity demand, but also effectively simulate the spatial changes of land use under multiple scenarios. It fully leverages the advantages of the two models in terms of quantity prediction and spatial distribution, and realizes the dual simulation of land use in both space and quantity.

[0049] Step S105: Construct a multi-scenario dynamic ecological security pattern based on the future land use raster map.

[0050] It should be noted that the multi-scenario dynamic ecological security pattern refers to a series of spatiotemporally coupled and scenario-differentiated ecological security pattern maps generated based on future land use grid maps under multiple scenarios and time nodes, through methods such as landscape ecological risk assessment, ecological source area identification, and ecological corridor construction.

[0051] In one optional embodiment, a multi-scenario dynamic ecological security pattern is constructed based on a future land use raster map, specifically including the following steps: Ecological source areas are identified based on future land use raster maps and multi-source environmental data of the target watershed, resulting in a binary map of the ecological source areas. A corrected ecological resistance coefficient raster map is determined based on the future land use raster maps and multi-source environmental data of the target watershed. Ecological corridors and strategic nodes are identified based on the binary map of the target ecological source areas and the corrected ecological resistance coefficient raster map.

[0052] In one specific embodiment, ecological source areas are identified based on a future land use raster map and multi-source environmental data of the target watershed, resulting in a binary map of the ecological source areas, including: The future land use raster map is converted into a binary map, which includes a foreground region and a background region. Morphological operations on the foreground region are performed using the eight-neighborhood algorithm to determine the core area, which is then identified as a candidate ecological source area. Spatial distribution maps of at least one ecosystem service indicator are obtained based on multi-source environmental data from the target watershed. The weights of each ecosystem service indicator are determined using the analytic hierarchy process (AHP), and the spatial distribution maps are weighted and overlaid to generate an ecosystem importance score map. The comprehensive ecosystem importance score map is then spatially masked and overlaid with the core area to select regions within the core area whose ecosystem importance scores are higher than a preset threshold. Based on these regions, a binary map of the target ecological source area is generated.

[0053] In a more specific embodiment, converting the future land use raster map into a binary map includes setting the ecological land use category in the future land use raster map to 1, and the rest to 0. Here, 1 represents a foreground pixel, and 2 represents a background pixel.

[0054] For example, the foreground includes woodlands, grasslands, wetlands, and water bodies, while the background includes built-up land.

[0055] In one alternative embodiment, morphological operations include erosion, dilation, opening, and closing operations.

[0056] In one alternative embodiment, the landscape category includes at least one of the following categories: core area, bridging area, island patch, edge area, pore, loop area, and branch line, etc.

[0057] In one alternative embodiment, the ecosystem service indicators include at least one of the following: water conservation, soil retention, biodiversity, and solid oxygen release.

[0058] In one specific embodiment, water conservation indicators are determined using a water balance model.

[0059] In a more specific embodiment, the expression for the water balance model is: (2); Where W represents the average water conservation capacity, P represents the average rainfall, and E represents the average evaporation.

[0060] In another specific embodiment, soil retention indicators are determined using the RUSLE (Revised Universal Soil Loss Equation) model.

[0061] In a more specific embodiment, the expression for the RUSLE model is: (3); Where A represents the average annual soil retention per unit area, and R represents the rainfall erosivity factor. LS represents the soil erodibility factor, C represents the cover and management factor, and P represents the soil and water conservation measures factor.

[0062] In another specific embodiment, the carbon sequestration and oxygen release index is determined using the CASA (Carnegie-Ames-Stanford Approach) model.

[0063] In a more specific embodiment, the expression for the CASA model is: (4); in, This represents the net primary productivity of vegetation at spatial location x during time period t. This represents the photosynthetically active radiation absorbed by vegetation at spatial location x during time period t. This represents the light energy conversion rate of vegetation at spatial location x during time period t.

[0064] In another specific embodiment, biodiversity indicators are determined using the InVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) model.

[0065] In another optional embodiment, the modified ecological resistance coefficient raster map is determined based on the future land use raster map and multi-source environmental data of the target watershed, specifically including the following steps: Based on future land use raster maps and multi-source environmental data of the target watershed, landscape structure index raster maps, landscape vulnerability index raster maps, and landscape loss index raster maps are generated. Based on the landscape structure index raster map, landscape vulnerability index raster map, and landscape loss index raster map, a landscape ecological risk index raster map is calculated. Based on the future land use raster map and local ecological planning standards, the basic ecological resistance coefficient corresponding to each land use type is determined. Based on the basic ecological resistance coefficient, a basic ecological resistance raster map is generated. At least one disaster-causing factor is obtained. Based on at least one disaster-causing factor, a geological hazard sensitivity raster map is generated. Based on the geological hazard sensitivity raster map and the future land use raster map, a geological hazard average sensitivity raster map is generated. Based on the basic ecological resistance raster map, landscape ecological risk index raster map, geological hazard sensitivity raster map, and geological hazard average sensitivity raster map, a corrected ecological resistance coefficient raster map is calculated.

[0066] This application, taking into account the characteristics of the watershed and the ecological significance of the landscape index, reconstructed the landscape fragmentation index, landscape separation index, and landscape dominance index through repeated experiments and calculations, by discarding the calculation methods of duplicate and extreme values. Furthermore, using landscape ecology methods, the landscape structure index, landscape vulnerability index, and landscape loss index were selected as risk assessment indicators to construct a landscape ecological risk index and correct the landscape ecological resistance coefficient.

[0067] In one specific embodiment, landscape fragmentation indices include edge density and area-weighted average shape factor. Edge density reveals the degree to which a landscape or type is divided by boundaries and is a direct reflection of the degree of landscape fragmentation; the higher the edge density, the higher the landscape fragmentation. Area-weighted average shape factor is one of the important indicators for measuring the complexity of landscape spatial patterns. For the shape analysis of natural patches or natural landscapes, it also has another significant ecological meaning, namely the so-called edge effect.

[0068] Landscape separation indices include patch assemblage index, aggregation index, and clustering index. The patch assemblage index describes the degree of clustering or extension trend of different patch types within a landscape; due to its spatial information content, it is one of the most important indices for describing landscape patterns. The aggregation index, derived from the calculation of the proximity matrix at the patch type level, is another important indicator reflecting the degree of landscape aggregation and separation. The clustering index significantly reflects the distribution characteristics of ecosystems severely constrained by certain natural conditions.

[0069] Landscape dominance indices include the maximum patch index, which helps determine the modalities or dominant types of a landscape. The value of the maximum patch index determines the ecological characteristics of the landscape, such as the abundance of dominant species and internal species. Changes in the value can alter the intensity and frequency of disturbances, reflecting the direction and strength of human activities.

[0070] The landscape structure index is a weighted sum of edge density, area-weighted average shape factor, patch assemblage index, aggregation index, clustering index, and maximum patch index. It is used to reflect the degree of disturbance and loss to the ecosystems represented by different landscapes. The weights are objectively assigned using a weighting method based on consistency, relevance, and criterion strength.

[0071] The landscape vulnerability index indicates the sensitivity of different landscape types to external disturbances; the higher the value, the greater the ecological risk. The magnitude of landscape vulnerability is related to its stage in the natural succession of the landscape and can be assigned a value based on the characteristics of the study area and experience. For example, landscape types can be assigned values ​​from highest to lowest vulnerability as follows: 7 bare land, 6 desert, 5 water bodies, 4 cultivated land, 3 grassland, 2 forest land, and 1 construction land.

[0072] The landscape loss index indicates the degree of loss of natural attributes of ecosystems represented by different landscape types when they are disturbed by external factors. It is expressed as the geometric mean of the landscape structure index and the landscape vulnerability index.

[0073] The expression for the landscape ecological risk index is: (5); in, This represents the landscape ecological risk index of the k-th grid. This represents the area of ​​the i-th land use category within the k-th grid. This represents the total area of ​​the k-th grid cell. denoted by , represents the landscape loss index of the i-th land use category, and n represents the number of land use categories included in a grid.

[0074] Geological disasters not only alter the original topography and landforms of a watershed but also affect its ecosystem services, hindering existing ecological corridors, disrupting landscape connectivity, and creating habitat isolation. Therefore, it is necessary to modify ecological resistance surfaces from a geological disaster perspective to improve the rationality and accuracy of subsequent ecological corridor identification. Influenced by natural factors such as geology, landforms, and climate, geological disasters can include landslides and debris flows. These disasters are all rock and soil displacement disasters and are closely related to the properties of the rock and soil, slope, rainfall, and vegetation cover.

[0075] In one specific embodiment, the properties of soil and rock, slope, rainfall, and vegetation cover are selected as disaster-causing factors. After normalization, they are superimposed with equal weights to evaluate the geological disaster sensitivity of the Kaikong River Basin, and the basic ecological resistance surface is modified based on this.

[0076] In one specific embodiment, based on the basic ecological resistance raster map, the landscape ecological risk index raster map, the geological hazard sensitivity raster map, and the average geological hazard sensitivity raster map, the modified ecological resistance coefficient raster map is calculated using the following formula, including: (6); Where R represents the basic ecological resistance coefficient, which is determined by land use type. This represents the landscape ecological risk index of the k-th grid. Indicates the landscape ecological risk adjustment coefficient. This represents the geological hazard sensitivity of the k-th grid. This indicates the average geological hazard sensitivity of land use category a; (7); Where m represents the number of grid cells belonging to land use category a.

[0077] It should be noted that, It was set based on experience.

[0078] In another optional embodiment, ecological corridors and strategic nodes are identified based on a binary map of the target ecological source area and a modified raster map of ecological resistance coefficients, including: The spatial distribution of primary ecological corridors is constructed based on the binary map of the target ecological source area and the modified ecological resistance coefficient raster map; a current density raster map is generated based on the spatial distribution of primary ecological corridors; and ecological strategic nodes are determined based on the current density raster map.

[0079] It should be noted that ecological corridors are an important component of a complete ecosystem of mountains, rivers, forests, fields, lakes, grasslands, and deserts. They are distributed linearly or in a strip-like pattern in the ecological environment, serving to connect spatially isolated and scattered habitat patches. Ecological strategic nodes include pinch points and obstruction points. Pinch points refer to points in ecological corridors with high landscape mobility, while obstruction points refer to points that hinder connectivity.

[0080] In a more specific embodiment, ecological corridors and strategic nodes in the watershed are identified based on circuit theory. The Linkage Mapper tool (connectivity analysis toolbox) combines circuit theory and motion ecology through the random walk characteristics of charge. Ecological source areas and corrected resistance surface data are imported to create a mapping of minimum-cost channels, thereby identifying ecological corridors. Based on the creation of ecological corridors, a cumulative current density map is generated by calling the open-source program Circuitscape (circuit landscape analysis software) to identify ecological strategic points.

[0081] Step S106: Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, obtain the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations.

[0082] It should be noted that the driving factors in the second set of driving factors are those corresponding to the future land use raster map. A geographic detector is used to analyze the impact of each driving factor on the watershed's landscape ecological security pattern. The geographic detector is a novel statistical method for detecting spatial heterogeneity and revealing its underlying driving factors. This method makes no linear assumptions and has an elegant form and clear physical meaning.

[0083] In one alternative embodiment, a geographic detector is used to detect the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driver combinations.

[0084] In one specific embodiment, the process of obtaining a ranking table of driving factor contributions, a matrix of factor interaction types, and identification results of key risk driving combinations based on a multi-scenario dynamic ecological security pattern and a second set of driving factors includes: determining the dependent variable based on the multi-scenario dynamic ecological security pattern, using the second driving factor as the independent variable, conducting detection and analysis through a geographic detector, and obtaining a ranking table of driving factor contributions, a matrix of factor interaction types, and identification results of key risk driving combinations.

[0085] In one specific embodiment, the ranking table of driving factor contributions is calculated using the q-statistic (factor detector) in the geographic detector, where the expression for q is: (8); Where q represents the factor explanatory power index, and L represents the number of categories. This represents the number of raster cells in the h-th factor category. Let N represent the variance of the dependent variable in the h-th factor category, and N represent the number of grid cells. SSW represents the population variance of the dependent variable across all raster counts. SST= .

[0086] In one specific embodiment, the factor interaction type matrix reveals whether the impact on ecological risk is enhanced or weakened when any two factors act together.

[0087] In a more specific embodiment, five interaction types are included: Nonlinear reduction: q(x1∩x2) <min(q(x1), q(x2)) Single-factor nonlinear reduction: Min(q(x1), q(x2)) <q(x1∩x2)<Max(q(x1),q(x2)) Two-factor enhancement: q(x1∩x2)>Max(q(x1), q(x2)) Independent: q(x1∩x2) = q(x1) + q(x2) Nonlinear enhancement: q(x1∩x2)>q(x1)+q(x2) In one specific embodiment, the key risk-driven portfolio identification results include t-statistics and F-statistics.

[0088] In a more specific embodiment, the t-statistic is calculated using the following formula: (9); in, This represents the mean of the dependent variable Y in category h=1. This represents the mean of the dependent variable Y in category h=2. This represents the variance of the dependent variable Y in class h=1. This represents the variance of the dependent variable Y in class h=2. This indicates the number of rasters in category h=1. represents the number of rasters in category h=2, and t represents the statistic; The F-statistic is calculated using the following formula: (10); in, The number of grid cells represents the driving factor x1. The number of grid cells represents the driving factor x2. This represents the SSW corresponding to the driving factor x1. This represents the SSW corresponding to the driving factor x2.

[0089] Step S107: Determine the target ecological restoration protected area based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations.

[0090] In one optional embodiment, the target ecological restoration protected area is determined based on a ranking table of driving factor contributions, a matrix of factor interaction types, and the identification results of key risk driving combinations. This includes: determining target driving factors based on the ranking table of driving factor contributions; obtaining spatial raster maps corresponding to the target driving factors; identifying target factor combinations based on the matrix of factor interaction types; overlaying the spatial raster maps corresponding to the target driving factors with the target factor combinations to obtain a collaborative risk raster map; and determining the target ecological restoration protected area based on the identification results of key risk driving combinations and the collaborative risk raster map.

[0091] In one specific embodiment, the top 3–5 high-contribution factors are determined based on a ranking table of driving factor contributions, such as road density (q=0.71), nighttime light index (q=0.63), and cultivated land ratio (q=0.52). Continuous factors can also be classified, for example, road density: low road density (road density less than 1), medium road density (road density 1–3), and high road density (road density greater than 3). Nighttime light index includes weak nighttime light index (nighttime light index less than 5), medium nighttime light index (nighttime light index 5–15), and strong nighttime light index (nighttime light index greater than 15); a high-risk level layer is generated for each factor; "dual-factor enhancement" or "non-linear enhancement" types are identified; spatial intersection is calculated between "high-density road areas" and "strong nighttime light areas"; the average ERI (landscape ecological risk index) of the intersection area is calculated, and if it is significantly higher than the background value, it is confirmed as a "synergistic risk hotspot"; the discrimination rules for "key combinations" are clearly defined, and target ecological restoration protection areas are delineated.

[0092] Step S108: Generate restoration strategies for the target ecological restoration protected area.

[0093] In one alternative embodiment, the restoration strategy may include returning farmland to grassland / forest, rebuilding farmland shelterbelts, and restoring ecological corridors, etc. This application does not limit the specific content of the restoration strategy.

[0094] Corresponding to the method for generating restoration strategies for watershed landscape ecological protection zones provided in this application, this application also provides an apparatus for generating restoration strategies for watershed landscape ecological protection zones, such as... Figure 2 As shown, the device for generating restoration strategies for watershed landscape ecological protection zones includes: The acquisition module 201 is used to acquire multi-period historical land use raster maps of the target watershed and determine the base year land use raster map from the historical land use raster maps. The first determining module 202 is used to determine the transfer probability matrix based on the historical land use raster map; The first prediction module 203 is used to predict multi-scenario area demand data for the target watershed based on the actual area vectors and transition probability matrices of each land use category extracted from the land use raster map of the base year. The second prediction module 204 is used to predict the future land use raster map based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, through the FLUS model. The first set of driving factors has the same resolution and the same range as the base year land use raster map. Module 205 is used to construct a multi-scenario dynamic ecological security pattern based on the future land use raster map; The second determining module 206 is used to obtain a ranking table of driving factor contribution, a matrix of factor interaction types, and the identification results of key risk driving combinations based on the multi-scenario dynamic ecological security pattern and the second set of driving factors. The second set of driving factors is the driving factors corresponding to the future land use raster map. The third determination module 207 is used to determine the target ecological restoration protected area based on the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The generation module 208 is used to generate restoration strategies for the target ecological restoration protected area.

[0095] Corresponding to the method for generating restoration strategies for watershed landscape ecological protection zones provided in this application, this application also provides an electronic device for executing the method for generating restoration strategies for watershed landscape ecological protection zones, such as... Figure 3 As shown, the electronic device includes: a processor 301; and a memory 302 for storing a program for generating a watershed landscape ecological protection zone restoration strategy. After the device is powered on and the program for generating the watershed landscape ecological protection zone restoration strategy is run through the processor, the following steps are performed: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps; Determine the transition probability matrix based on historical land use raster maps; Based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, multi-scenario area demand data for the target watershed are predicted. Based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, the future land use raster map is predicted by the FLUS model. The first set of driving factors has the same resolution and range as the base year land use raster map. Constructing a dynamic ecological security pattern based on future land use grid maps; Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, we obtained the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The second set of driving factors is the driving factor corresponding to the future land use raster map. The target ecological restoration protected area was determined based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations. Generate restoration strategies for the target ecological restoration protected area.

[0096] Corresponding to the method for generating restoration strategies for watershed landscape ecological protection zones provided in this application, this application also provides a computer-readable storage medium storing a program for generating restoration strategies for watershed landscape ecological protection zones. This program is executed by a processor to perform the following steps: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps; Determine the transition probability matrix based on historical land use raster maps; Based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, multi-scenario area demand data for the target watershed are predicted. Based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, the future land use raster map is predicted by the FLUS model. The first set of driving factors has the same resolution and range as the base year land use raster map. Constructing a dynamic ecological security pattern based on future land use grid maps; Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, we obtained the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The second set of driving factors is the driving factor corresponding to the future land use raster map. The target ecological restoration protected area was determined based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations. Generate restoration strategies for the target ecological restoration protected area.

[0097] Corresponding to the watershed landscape ecological protection zone restoration strategy generation method provided in the embodiments of this application, the embodiments of this application also provide a computer program containing instructions, which, when executed by a computer, cause the computer to perform the following steps: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps; Determine the transition probability matrix based on historical land use raster maps; Based on the actual area vectors and transition probability matrices of each land use category extracted from the base year land use raster map, multi-scenario area demand data for the target watershed are predicted. Based on the base year land use raster map, the first set of driving factors, multi-scenario area demand data, neighborhood factors and probability transition matrix, the future land use raster map is predicted by the FLUS model. The first set of driving factors has the same resolution and range as the base year land use raster map. Constructing a dynamic ecological security pattern based on future land use grid maps; Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, we obtained the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The second set of driving factors is the driving factor corresponding to the future land use raster map. The target ecological restoration protected area was determined based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations. Generate restoration strategies for the target ecological restoration protected area.

[0098] It should be noted that for a detailed description of the watershed landscape ecological protection zone restoration strategy generation device, electronic device, computer-readable storage medium and computer program product provided in the embodiments of this application, please refer to the relevant description of the embodiments of the watershed landscape ecological protection zone restoration strategy generation method provided in the embodiments of this application, which will not be repeated here.

[0099] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

[0100] In a typical configuration, an electronic device includes one or more processors (Central Processing Units), input / output interfaces, network interfaces, and memory.

[0101] Memory may include non-persistent storage in computer-readable media, such as random access memory and / or non-volatile memory, like read-only memory or flash memory. Memory is an example of computer-readable media.

[0102] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable operations, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PCM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital video disc (DMCD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include non-transitory computer-readable media, such as modulated data signals and carrier waves.

[0103] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, compact disc read-only memory, optical storage, etc.) containing computer-usable program code.

[0104] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

Claims

1. A method for generating restoration strategies for watershed landscape ecological protection zones, characterized in that, include: Obtain multi-period historical land use raster maps of the target watershed, and determine the base year land use raster map from the historical land use raster maps; Based on the historical land use raster map, determine the transfer probability matrix; Based on the actual area vectors of each land use category extracted from the land use raster map of the base year and the transition probability matrix, the multi-scenario area demand data of the target watershed is predicted. Based on the baseline year land use raster map, the first set of driving factors, the multi-scenario area demand data, the neighborhood factor, and the probability transition matrix, the future land use raster map is predicted using the FLUS model, wherein the first set of driving factors has the same resolution and the same range as the baseline year land use raster map. A multi-scenario dynamic ecological security pattern is constructed based on the aforementioned future land use grid map; Based on the multi-scenario dynamic ecological security pattern and the second set of driving factors, a ranking table of driving factor contribution, a matrix of factor interaction types, and the identification results of key risk driving combinations are obtained. The second set of driving factors is the driving factor corresponding to the future land use raster map. The target ecological restoration protected area is determined based on the ranking table of driving factor contributions, the matrix of factor interaction types, and the identification results of key risk driving combinations. Generate restoration strategies for the target ecological restoration protected area.

2. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 1, characterized in that, The determination of the migration probability matrix based on the historical land use raster map includes: For any two periods of the historical land use raster map, count the number of pixels for the initial land use category and the number of pixels for the final land use category in the historical land use raster map; The transition probability is calculated based on the number of pixels in the initial land use category and the number of pixels in the final land use category. A transition probability matrix is ​​generated based on the transition probabilities.

3. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 1, characterized in that, The construction of a multi-scenario dynamic ecological security pattern based on the future land use raster map includes: Based on the future land use raster map and the multi-source environmental data of the target watershed, ecological source areas are identified, and a binary map of ecological source areas is obtained. Based on the future land use raster map and the multi-source environmental data of the target watershed, a modified ecological resistance coefficient raster map is determined. Ecological corridors and strategic nodes are identified based on the binary map of the target ecological source area and the modified ecological resistance coefficient raster map.

4. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 3, characterized in that, The process of identifying ecological source areas based on the future land use raster map and multi-source environmental data of the target watershed, and obtaining a binary map of the ecological source areas, includes: The future land use raster map is converted into a binary map, which includes a foreground area and a background area; The core region is determined by performing morphological operations on the foreground region using the eight-neighborhood algorithm, and the core region is identified as a candidate ecological source area. Based on the multi-source environmental data of the target watershed, obtain a spatial distribution map of at least one ecosystem service indicator; The weights of each ecosystem service indicator are determined using the analytic hierarchy process (AHP), and the spatial distribution map is weighted and superimposed using these weights to generate an ecosystem importance score map. The ecological importance comprehensive score map is spatially masked and overlaid with the core area to filter out areas in the core area whose ecological importance scores are higher than a preset score threshold, and a target ecological source area binary map is generated based on the areas.

5. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 3, characterized in that, The determination of the corrected ecological resistance coefficient raster map based on the future land use raster map and multi-source environmental data of the target watershed includes: Based on the future land use raster map and the multi-source environmental data of the target watershed, a landscape structure index raster map, a landscape vulnerability index raster map, and a landscape loss index raster map are generated. Based on the landscape structure index raster map, landscape vulnerability index raster map, and landscape loss index raster map, the landscape ecological risk index raster map is calculated. Based on the aforementioned future land use grid map and local ecological planning standards, the basic ecological resistance coefficient corresponding to each land use type is determined; A basic ecological resistance raster map is generated based on the aforementioned basic ecological resistance coefficient. Obtain at least one hazard-causing factor; Generate a geological hazard sensitivity raster map based on at least one of the aforementioned hazard-causing factors; An average geological hazard sensitivity raster map is generated based on the geological hazard sensitivity raster map and the future land use raster map; Based on the aforementioned basic ecological resistance raster map, landscape ecological risk index raster map, geological hazard sensitivity raster map, and average geological hazard sensitivity raster map, the corrected ecological resistance coefficient raster map is calculated.

6. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 1, characterized in that, The results of obtaining the driving factor contribution ranking table, factor interaction type matrix, and key risk driving combination identification results based on the multi-scenario dynamic ecological security pattern and the second driving factor set include: Based on the multi-scenario dynamic ecological security pattern, the dependent variable is determined, and the second driving factor is used as the independent variable. The geographic detector is used for detection and analysis to obtain the driving factor contribution ranking table, the factor interaction type matrix, and the key risk driving combination identification results.

7. The method for generating restoration strategies for watershed landscape ecological protection zones according to claim 1, characterized in that, The determination of target ecological restoration protected areas based on the driving factor contribution ranking table, factor interaction type matrix, and key risk driver combination identification results includes: The target driving factor is determined based on the driving factor contribution ranking table. Obtain the spatial raster map corresponding to the target driving factor; Identify target factor combinations based on the aforementioned factor interaction type matrix; Based on the combination of target factors, the spatial raster maps corresponding to the target driving factors are superimposed to obtain a collaborative risk raster map. The target ecological restoration protected area is determined based on the key risk-driven combination identification results and the collaborative risk raster map.

8. A device for generating restoration strategies for watershed landscape ecological protection zones, characterized in that, include: The acquisition module is used to acquire multi-period historical land use raster maps of the target watershed and determine the base year land use raster map from the historical land use raster maps; The first determining module is used to determine the transfer probability matrix based on the historical land use raster map; The first prediction module is used to predict multi-scenario area demand data for the target watershed based on the actual area vectors of each land use category extracted from the land use raster map of the base year and the transition probability matrix. The second prediction module is used to predict the future land use raster map based on the base year land use raster map, the first set of driving factors, the multi-scenario area demand data, the neighborhood factor and the probability transition matrix, using the FLUS model. The first set of driving factors has the same resolution and the same range as the base year land use raster map. The construction module is used to construct a multi-scenario dynamic ecological security pattern based on the future land use raster map; The second determining module is used to obtain a ranking table of driving factor contribution, a matrix of factor interaction types, and the identification result of key risk driving combination based on the multi-scenario dynamic ecological security pattern and the second set of driving factors. The second set of driving factors is the driving factor corresponding to the future land use raster map. The third determination module is used to determine the target ecological restoration protected area based on the driving factor contribution ranking table, the factor interaction type matrix and the key risk driving combination identification results. The generation module is used to generate restoration strategies for the target ecological restoration protected area.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the watershed landscape ecological protection zone restoration strategy generation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the watershed landscape ecological protection zone restoration strategy generation method according to any one of claims 1-7.

Images