Spatial partitioning method for measuring terrain fragmentation

Abstract

The invention discloses a spatial partitioning method for measuring terrain fragmentation, and the method comprises the steps: obtaining DEM and county-level and district-level administrative division data in a given range, and obtaining a plurality of spatial units; for each space unit, according to the average gradient, the elevation variation and the drainage basin density combination of the landform, obtaining an overall fragmentation feature; obtaining positive and negative topographic features according to the combination of the positive topographic proportion and the negative topographic proportion of the landform; gradient distribution characteristics are obtained according to the combination of the gradient area ratios of 2-5 degrees, 5-15 degrees, 15-35 degrees and 35-90 degrees of the landform; establishing a terrain fragmentation measurement system according to the overall fragmentation characteristics, the positive and negative terrain characteristics and the gradient distribution characteristics; calculating a weight by using an entropy weight method and obtaining a comprehensive evaluation value; a natural breakpoint method is adopted to classify the fragmentation comprehensive evaluation values, and spatial distribution of different types of terrain fragmentation degrees is obtained; and based on a spatial filtering thought, judging whether the category of each spatial unit is the same as the main category of all adjacent units one by one, and if not, resetting the category of the spatial unit as the main category corresponding to the neighborhood unit of the spatial unit. The step can be repeated to obtain a spatial partitioning result in which the same category is continuously distributed and different categories are not crossed. According to the method, the terrain fragmentation features can be objectively and comprehensively measured, then an automatic partitioning result and key influence indexes are formed, and the problems of subjective judgment or single inspection dimension and the like which are likely to occur in a traditional method are solved.

Description

Technical Field

[0001] The present invention belongs to the fields of geographic information technology, digital terrain analysis, resource and environmental technology, and relates to a spatial partitioning method for measuring terrain fragmentation, and in particular to a spatial automatic partitioning method for county unit spatial filtering by comprehensively evaluating terrain fragmentation characteristics. Background Art

[0002] Terrain fragmentation is the core research object in the field of resource and environmental technology. Fragmentation is a process and state in which the overall space tends to split and become discontinuous under the influence of natural or human factors. It is an important reason for poor regional connectivity, lagging traffic development, difficulty in resource development, limited biological activities, and low degree of agricultural mechanization. Fragmentation can not only represent the overall folding degree of topography at the macro scale, but also depict the degree of local surface segmentation at the micro scale. The study of fragmentation is not only conducive to the expansion of topographic and geomorphological theory, but also has important reference value in the fields of transportation layout, resource development, and disaster assessment. Terrain fragmentation makes spatial communication difficult within the Loess Plateau, and to a certain extent determines the form and scale of human activities, resulting in regional development facing prominent problems such as serious population loss, low productivity, and severe soil erosion, which seriously affects and restricts the layout and development of road traffic, social economy, agricultural production, and resource protection and development. Therefore, studying the degree of terrain fragmentation and spatial differentiation in mountainous areas can provide an important basis for regional transportation resilience evaluation, production layout optimization, and ecological environmental conservation.

[0003] Previous studies have focused on the influence of traditional terrain factors such as elevation, slope, slope aspect, slope length, surface cutting depth, and surface undulation on surface morphological changes and regional development. At present, the main methods used to study terrain morphological characteristics are: statistical index method, fractal index method, and point / line / surface characteristics, as well as information spectrum analysis method, slope spectrum analysis method, terrain texture analysis method, etc. There are also a few studies that use EEMD (Ensemble Empirical Mode Decomposition) and other methods to analyze terrain characteristics. The above research methods can generate a variety of terrain characteristics, but none of them fully characterizes and considers terrain fragmentation. However, a small number of studies use elevation standard deviation and elevation coefficient of variation to characterize terrain fragmentation. For example, the elevation coefficient of variation obtained by the standard deviation of elevation in different regions within each county and the ratio of the standard deviation of elevation in the region to the mean represents the overall degree of terrain fragmentation in the region. Although previous studies have measured terrain fragmentation from a few aspects, they have not comprehensively characterized terrain fragmentation, and have not further constructed a more comprehensive terrain fragmentation measurement method. In addition, some studies have proposed using spatial segmentation methods to explore topographic and geomorphic features, such as using remote sensing images to fuse terrain features for image depth segmentation, automatically segmenting terrain and geomorphic data based on digital elevation model (DEM) data to extract spatial units, and using the threshold segmentation of terrain positive and negative openings and their difference images to extract topographic and geomorphic ditch lines. Although these segmentation methods can divide spatial units according to terrain similarity and have a certain explanation for the concept of terrain fragmentation, they have not explored the spatial heterogeneity of terrain fragmentation. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a spatial partitioning method for measuring terrain fragmentation, which can automatically generate terrain fragmentation partitioning results and further obtain key influencing indicators, so as to avoid the problems in traditional methods that cannot take into account the multi-dimensional considerations of terrain fragmentation, the single indicator type and the need for manual judgment, the arbitrary technical results and low work efficiency.

[0005] Technical solution: To achieve the above purpose, the spatial partitioning method for measuring terrain fragmentation described in the present invention comprises the following steps:

[0006] S1: Obtain DEM and county-level administrative division data within a given range to obtain multiple spatial units;

[0007] S2: For each spatial unit, the overall fragmentation characteristics are obtained according to the combination of the average slope, elevation variation, and watershed density of the topography; the positive and negative terrain characteristics are obtained according to the combination of the positive terrain proportion and the negative terrain proportion of the topography; the slope distribution characteristics are obtained according to the combination of the 2°-5°°, 5°-15°, 15°-35°°, and 35°-90° slope area ratios of the topography;

[0008] S3: Establish a terrain fragmentation measurement system based on overall fragmentation characteristics, positive and negative terrain characteristics, and slope distribution characteristics;

[0009] S4: Use the entropy weight method to calculate the weight and obtain the comprehensive evaluation value;

[0010] S5: Use the natural breakpoint method to classify the degree of terrain fragmentation based on the comprehensive evaluation value of fragmentation, and obtain the spatial distribution of different categories;

[0011] S6: Based on the idea of ​​spatial filtering, determine whether the category of each spatial unit is the same as the main category of all adjacent units. If not, reset the category of the spatial unit to the main category corresponding to its neighboring units. This step can be repeated to obtain a spatial partition result with continuous distribution of the same category and no overlap of different categories.

[0012] In a preferred embodiment, the method further comprises: performing a correlation analysis on the selected terrain spatial distribution characteristic indicators, and outputting key influencing indicators of terrain fragmentation spatial partitioning.

[0013] In a preferred embodiment, the spatial unit is a vector spatial unit formed based on the county range.

[0014] In a preferred embodiment, the average slope of the spatial unit is calculated by using the partition statistics tool of geographic information related analysis software on the DEM data of the study area, and the specific calculation is:

[0015] slope:

[0016] in: and The slope is the rate of change of elevation (i.e., elevation gradient) along the X direction (east-west) and the Y direction (north-south). i is the slope in radians, which can be converted to percentage or degrees.

[0017] Average slope:

[0018] Where: Slope i is the slope value of each pixel in the i-th spatial unit; SlopeM iis the average slope value of all pixels in the i-th spatial unit; n represents the number of all pixels in the spatial unit.

[0019] In a preferred embodiment, the elevation variation is the standard deviation of all pixel elevation values ​​within the spatial unit, specifically:

[0020] Elevation variation:

[0021] Where: j is the standard deviation of the elevation of the jth spatial unit; N is the total number of pixels (grid units); Z i is the elevation value of the i-th pixel.

[0022] μ is the average value of all pixel elevation values, specifically:

[0023]

[0024] In a preferred embodiment, the watershed density is the ratio of the number of watersheds in a spatial unit divided by DEM to the total area of ​​the region, specifically:

[0025] Watershed density:

[0026] Where: D is the watershed density, usually in km / km 2 ; L is the total length of the main river channel in the basin; A is the total area of ​​the basin.

[0027] In a preferred embodiment, the positive terrain ratio is the ratio of the number of pixels in the spatial unit that are higher than the reference elevation value to the total number of pixels in the spatial unit, specifically:

[0028] Benchmark elevation: Binwidth = (E max -E min ) / n

[0029] Where E represents the elevation value of the pixel; E max Indicates the maximum elevation value of the pixel in the spatial unit; E min Indicates the maximum elevation value of the pixel in the spatial unit; n indicates the number of grouping intervals used to generate the histogram of the elevation value of a spatial unit. The benchmark elevation is to generate a histogram of the elevation value of a spatial unit, using n grouping intervals, calculating the elevation range of the main distribution interval (maximum value minus minimum value), and then calculating the proportion of the number of pixels below this range.

[0030] Positive terrain ratio: SB i =S E>Binwidth / S i

[0031] Among them, SB iIt represents the ratio of the pixel area above the benchmark elevation to the total pixel area in the i-th spatial unit; S E>Binwidth represents the pixel area above the benchmark elevation in the i-th spatial unit; S i Represents the area of ​​all pixels in the i-th spatial unit.

[0032] In a preferred embodiment, the negative terrain ratio is the ratio of the number of pixels in the spatial unit that are lower than the reference elevation value to the total number of pixels in the spatial unit, specifically:

[0033] Negative terrain ratio: SH i =S E<Binwidth / S i

[0034] Among them, SH i It represents the ratio of the pixel area below the benchmark elevation to the total pixel area in the i-th spatial unit; S E<Binwidth represents the pixel area below the benchmark elevation in the i-th spatial unit; S i Represents the area of ​​all pixels in the i-th spatial unit.

[0035] In a preferred embodiment, the 2°-5° slope area ratio is the ratio of the number of pixels with a slope within 2°-5° to the total number of pixels in the spatial unit, specifically:

[0036] 2°-5° Slope Area Ratio: SA 0205i =S 0205i / S i

[0037] Among them: SA 0205i It represents the ratio of the number of pixels with slope within 2°-5° in the i spatial unit to the total number of pixels in the spatial unit; S 0205i Indicates the number of pixels with slopes within 2°-5° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit.

[0038] In a preferred embodiment, the 5°-15° slope area ratio is the ratio of the number of pixels with a slope within 5°-15° to the total number of pixels in the spatial unit, specifically:

[0039] 5°-15° Slope Area Ratio: SA 0515i =S 0515i / S i

[0040] Among them: SA 0515i It represents the ratio of the number of pixels with slopes within 5°-15° in the i spatial unit to the total number of pixels in the spatial unit; S 0515i represents the number of pixels with slopes within 5°-15° in the i spatial unit; Si Represents the total number of pixels in the i spatial unit.

[0041] In a preferred embodiment, the 15°-35° slope area ratio is the ratio of the number of pixels with a slope within 15°-35° to the total number of pixels in the spatial unit, specifically:

[0042] 15°-35° Slope Area Ratio: SA 1535i =S 1535i / S i

[0043] Among them: SA 1535i It represents the ratio of the number of pixels with slopes between 15° and 35° in the i spatial unit to the total number of pixels in the spatial unit; S 1535i Indicates the number of pixels with slopes between 15° and 35° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit.

[0044] In a preferred embodiment, the 35°-90° slope area ratio is the ratio of the number of pixels with a slope within 35°-90° to the total number of pixels in the spatial unit, specifically:

[0045] 35°-90° Slope Area Ratio: SA 3590i =S 3590i / S i

[0046] Among them: SA 3590i It represents the ratio of the number of pixels with slopes between 35° and 90° in the i spatial unit to the total number of pixels in the spatial unit; S 3590i Indicates the number of pixels with slopes between 35° and 90° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit.

[0047] In a preferred embodiment, the indicators are classified according to the expression of different aspects of the terrain, and the terrain fragmentation indicators are summarized into three major aspects.

[0048] In a specific implementation scheme, the entropy weight method calculates the entropy weight of each indicator according to the dispersion degree of each indicator data, and then revises the entropy weight to a certain extent according to each indicator, determines the objective indicator weight, and obtains the comprehensive evaluation value of terrain fragmentation, which is specifically:

[0049] Positive indicator normalization:

[0050] Negative indicator normalization:

[0051] Weight calculation:

[0052] Among them, m and n represent m samples and n indicators respectively; X ij Indicates the jth indicator and the specific value of the i-th sample; represents the jth index, the value of the i-th sample after standardization; max(X 1j ,……,X mj ) represents the maximum sample value in the jth index; min(X 1j ,……,X mj ) represents the smallest sample value in the jth index; P ij Represents the weight coefficient of the jth indicator.

[0053] In a preferred embodiment, the spatial distribution of the degree of terrain fragmentation is a comprehensive evaluation value of terrain fragmentation based on the natural breakpoint method of GIS-related software classification to visualize the categories to which each spatial unit in the study area belongs. The spatial filtering is calculated using the values ​​of the spatial unit area.

[0054] In a preferred embodiment, the steps of screening and determining key indicators include: performing correlation analysis on the data in the summary table of all selected terrain fragmentation indicators to obtain a correlation matrix diagram; according to the principle of multicollinearity, comparing the absolute values ​​of each indicator in the obtained correlation matrix diagram, screening indicators with absolute values ​​of correlation coefficients higher than a certain threshold (the threshold in this example is set to 0.75), and selecting one of the representative indicators as the key influencing indicator in terrain fragmentation.

[0055] In a preferred embodiment, after obtaining the cluster number corresponding to each spatial unit in step (6), the spatial units with the same cluster number are filled with a color, and the structure of the terrain fragmentation spatial partition is displayed in a visual way of colored blocks.

[0056] A spatial partitioning method for measuring terrain fragmentation according to the present invention comprises:

[0057] The spatial unit acquisition module is used to obtain the basic data of terrain fragmentation indicators within a given range. The basic data of terrain fragmentation indicators include DEM raster data of various terrain fragmentation indicators within the county vector range. The DEM raster data is the basic data for calculating various indicators.

[0058] Terrain fragmentation index measurement system establishment module: used to generate a terrain fragmentation index summary table for each terrain fragmentation index according to the calculation of multiple county unit terrain fragmentation indicators; the terrain fragmentation indicators include the average slope, elevation variation, and watershed density of the overall terrain fragmentation characteristics, the positive terrain ratio and negative terrain ratio of the positive and negative terrain characteristics, and the slope distribution characteristics of 2°-5°, 5°-15°, 15°-35°, and 35°-90° slope area ratios.

[0059] Entropy weight evaluation module: used to determine the weight coefficient of terrain fragmentation indicators. First, standardize and normalize the data of each indicator; after preliminary data processing, calculate the information entropy and information utility value of each indicator, and normalize each indicator; finally, calculate the weight coefficient of the indicator.

[0060] Spatial filtering clustering evaluation module: Based on the idea of ​​spatial filtering, it determines whether the category of each spatial unit is the same as the main category of all adjacent units. If not, the category of the spatial unit is reset to the main category corresponding to its neighboring units. This step can be repeated to obtain a spatial partition result with continuous distribution of the same category and no overlap of different categories.

[0061] The spatial partitioning method for measuring terrain fragmentation described in the present invention includes at least one computer device, wherein the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded into the processor, the method for automatic spatial partitioning of terrain fragmentation is implemented.

[0062] Beneficial effects: The present invention proposes a spatial partitioning method for measuring terrain fragmentation, which takes the geographic spatial morphology as a whole, comprehensively considers the multi-dimensional morphological characteristics, uses computer algorithms, and performs clustering operations based on the county spatial morphological characteristics, which can achieve the maximum approximation of the "comprehensive" rather than "single" measurement of the geographic spatial morphology; the output of the partitioning results is also based on the spatial automatic partitioning program, and the results are automatically converged to the state with the most distribution of clustering results, avoiding the randomness of technical results caused by manual interpretation. And further determine the key influencing indicators in the spatial partitioning of terrain fragmentation, which is conducive to grasping the characteristics of the geographic spatial morphological indicators in key areas in the planning and design within the research scope, and then grasping the geographic spatial morphological characteristics of the research scope as a whole. The present invention avoids the problems in the traditional method that the multi-dimensional consideration of terrain fragmentation cannot be taken into account, the indicator type is single and requires manual judgment, the technical results are very random, and the work efficiency is low. It provides a reliable rational scientific basis for being able to analyze the internal mechanism of geographic spatial morphology and then explore the structural laws of geographic space. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 The present invention is a flowchart of the method implemented in the present invention.

[0064] Figure 2 This is the spatial distribution diagram of the indicators preliminarily selected for the present invention.

[0065] Figure 3 This is a correlation matrix diagram of all indicators selected for the present invention.

[0066] Figure 4 For the screening of the indicators of the present invention, the correlation matrix diagram of the multicollinearity indicators is removed.

[0067] Figure 5 This is a diagram showing the comprehensive evaluation results of terrain fragmentation in the present invention.

[0068] Figure 6 This is the spatial partition result diagram obtained after the spatial filtering idea is adopted in the present invention. DETAILED DESCRIPTION

[0069] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0070] As shown in FIG. S1 , the embodiment of the present invention discloses a spatial partitioning method for measuring terrain fragmentation, which mainly includes the following steps:

[0071] S1: Obtain DEM and county-level administrative division data within a given range to obtain multiple spatial units.

[0072] In this step, the basic data of terrain fragmentation spatial morphology is obtained through the geospatial data cloud platform or other platforms, including 30m DEM data; the raster data is processed by geographic information related processing software, and combined with the study area contour data, the data that can be used for direct processing is obtained by clipping. The spatial unit data uses the county administrative boundary as the minimum spatial contour, and its contour is a closed polygonal space. The data format is generally shpefile format, and the marker of the study area class spatial unit is named after its county name.

[0073] S2: For each spatial unit, the overall fragmentation characteristics are obtained according to the combination of the average slope, elevation variation, and watershed density of the topography; the positive and negative terrain characteristics are obtained according to the combination of the positive terrain proportion and the negative terrain proportion of the topography; the slope distribution characteristics are obtained according to the combination of the 2°-5°, 5°-15°, 15°-35°, and 35°-90° slope area ratios of the topography.

[0074] In this step, the terrain fragmentation indicators include average slope, elevation variation, watershed density, positive terrain proportion, negative terrain proportion, and 2°-5°, 5°-15°, 15°-35°, and 35°-90° slope area ratios to represent the characteristics of the spatial unit. The area of ​​the spatial unit is obtained by calculating the geometric area within each closed contour.

[0075] The average slope is the mean of the slope values ​​of all pixels in the region; the elevation variation is the standard deviation of the elevation values ​​of all pixels in the region; the watershed density is the ratio of the number of watersheds in the region divided according to the DEM to the total area of ​​the region; the positive terrain ratio is the ratio of the number of pixels above the benchmark elevation value to the total number of pixels in the region; the negative terrain ratio is the ratio of the number of pixels below the benchmark elevation value to the total number of pixels in the region; the 2°-5° slope area ratio is the ratio of the number of pixels with a slope within 2°-5° to the total number of pixels in the region; the 5°-15° slope area ratio is the ratio of the number of pixels with a slope within 5°-15° to the total number of pixels in the region; the 15°-35° slope area ratio is the ratio of the number of pixels with a slope within 15°-35° to the total number of pixels in the region; and the 35°-90° slope area ratio is the ratio of the number of pixels with a slope within 35°-90° to the total number of pixels in the region.

[0076] S3: Establish a terrain fragmentation measurement system based on overall fragmentation characteristics, positive and negative terrain characteristics, and slope distribution characteristics.

[0077] In this step, generally at least two of the above indicators are selected to calculate the terrain fragmentation characteristic value of each spatial unit. In actual operation, the user can choose the combination by himself or give a default indicator combination, for example: the combination of average slope, elevation variation, and watershed density is used to comprehensively measure the overall fragmentation characteristics of terrain fragmentation; the combination of positive terrain proportion and negative terrain proportion is used to comprehensively measure the positive and negative terrain characteristics of terrain fragmentation; the combination of 2°-5°, 5°-15°, 15°-35° and 35°-90° slope area ratio is used to comprehensively measure the slope distribution characteristics of terrain fragmentation; the combination of all indicators "restores" the overall conceptual scope of the object of terrain fragmentation to the greatest extent, and comprehensively measures the comprehensive characteristics of terrain fragmentation of county spatial units.

[0078] S4: Use the entropy weight method to calculate the weight and obtain the comprehensive evaluation value.

[0079] In this step, the entropy weight method is used to first construct a judgment matrix for the evaluation index values ​​of each spatial unit calculated; secondly, the judgment matrix is ​​normalized to obtain a normalized judgment matrix; then, according to the definition of entropy, the information entropy of the evaluation index is calculated according to the evaluation index; finally, the weight value of the system is calculated. After obtaining the weight value, each spatial unit is comprehensively evaluated.

[0080] S5: The natural breakpoint method is used to classify the degree of terrain fragmentation based on the comprehensive evaluation value of fragmentation, and the spatial distribution of different categories is obtained.

[0081] In this step, according to the comprehensive evaluation value of each spatial unit obtained in step S4, the attribute table of the spatial unit in the shpefile file is spatially connected with the obtained comprehensive evaluation value according to the field in ArcGIS, and the comprehensive evaluation value field is classified using the natural breakpoint method to obtain 5 different categories for spatial distribution.

[0082] S6: Based on the idea of ​​spatial filtering, determine whether the category of each spatial unit is the same as the main category of all adjacent units. If not, reset the category of the spatial unit to the main category corresponding to its neighboring units. This step can be repeated to obtain a spatial partition result with continuous distribution of the same category and no overlap of different categories.

[0083] In this step, spatial distribution is performed based on the results of the comprehensive evaluation. According to the visualization results of the spatial distribution, spatial units with adjacent categories and dispersed distribution are divided into categories with concentrated distribution, so as to obtain spatial partitioning results with continuous distribution of the same categories and no overlap of different categories.

Claims

1. A spatial partitioning method for measuring terrain fragmentation, characterized in that: The steps include: S1: Obtain DEM and county-level administrative division data within a given range to obtain multiple spatial units; S2: For each spatial unit, the overall fragmentation characteristics are obtained based on the average slope, elevation variation, and watershed density of the topography; The positive and negative terrain characteristics are obtained by combining the proportion of positive terrain and the proportion of negative terrain in the topography; the slope distribution characteristics are obtained by combining the slope area ratios of 2°-5°, 5°-15°, 15°-35° and 35°-90° in the topography; S3: Establish a terrain fragmentation measurement system based on overall fragmentation characteristics, positive and negative terrain characteristics, and slope distribution characteristics; S4: Use the entropy weight method to calculate the weight and obtain the comprehensive evaluation value; S5: Use the natural breakpoint method to classify the degree of terrain fragmentation based on the comprehensive evaluation value of fragmentation, and obtain the spatial distribution of different categories; S6: Based on the idea of ​​spatial filtering, determine whether the category of each spatial unit is the same as the main category of all adjacent units. If not, reset the category of the spatial unit to the main category corresponding to its neighboring units. This step can be repeated to obtain a spatial partition result with continuous distribution of the same category and no overlap of different categories.

2. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The method further comprises: performing correlation analysis on the selected terrain spatial distribution characteristic indicators, and outputting key influencing indicators of terrain fragmentation spatial partitioning.

3. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The spatial unit is a vector spatial unit formed based on the county scope.

4. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The average slope of the i-th spatial unit is calculated by using the GIS-related software partition statistics tool on the DEM data of the study area, specifically: Slope: in: and The slope is the rate of change of elevation (i.e., elevation gradient) along the X direction (east-west) and the Y direction (north-south). i is the slope in radians, which can be converted to percentage or degrees. Average slope SlopeM i : Where: Slope i is the slope value of each pixel in the i-th spatial unit; SlopeM i is the average slope value of all pixels in the i-th spatial unit; n represents the number of all pixels in the spatial unit. The elevation variation is the standard deviation of all pixel elevation values ​​in the spatial unit, specifically: Elevation variation σ: Where: j is the standard deviation of the elevation of the jth spatial unit; N is the total number of pixels (grid units); Z i is the elevation value of the i-th pixel. μ is the average value of all pixel elevation values, specifically: The watershed density is the ratio of the number of watersheds in the spatial unit divided by DEM to the total area of ​​the region, specifically: Watershed density D: Where: D is the watershed density, usually in km / km 2 ; L is the total length of the main river channel in the basin; A is the total area of ​​the basin. The positive terrain ratio is the ratio of the number of pixels in the spatial unit that are higher than the benchmark elevation value to the total number of pixels in the spatial unit, specifically: Base elevation: Binwidth=(E max -E min ) / n Where E represents the elevation value of the pixel; E max Indicates the maximum elevation value of the pixel in the spatial unit; E min Indicates the maximum elevation value of the pixel in the spatial unit; n indicates the number of grouping intervals used to generate the histogram of the elevation value of a spatial unit. The benchmark elevation is to generate a histogram of the elevation value of a spatial unit, using n grouping intervals, calculating the elevation range of the main distribution interval (maximum value minus minimum value), and then calculating the proportion of the number of pixels below this range. Positive terrain ratio: SB i =S E>Binwidth / S i Among them, SB i It represents the ratio of the pixel area above the benchmark elevation to the total pixel area in the i-th spatial unit; S E>Binwidth represents the pixel area above the benchmark elevation in the i-th spatial unit; S i Represents the area of ​​all pixels in the i-th spatial unit. The negative terrain ratio is the ratio of the number of pixels below the reference elevation value in the spatial unit to the total number of pixels in the spatial unit, specifically: Negative terrain ratio: SH i =S E<Binwidth / S i Among them, SH i It represents the ratio of the pixel area below the benchmark elevation to the total pixel area in the i-th spatial unit; S E<Binwidth represents the pixel area below the benchmark elevation in the i-th spatial unit; S i Represents the area of ​​all pixels in the i-th spatial unit. The 2°-5° slope area ratio is the ratio of the number of pixels with a slope within 2°-5° to the total number of pixels in the spatial unit, specifically: IN 0205i =S 0205i / S i Among them: SA 0205i It represents the ratio of the number of pixels with slope within 2°-5° in the i spatial unit to the total number of pixels in the spatial unit; S 0205i Indicates the number of pixels with slopes within 2°-5° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit. The 5°-15° slope area ratio is the ratio of the number of pixels with a slope within 5°-15° to the total number of pixels in the spatial unit, specifically: IN 0515i =S 0515i / S i Among them: SA 0515i It represents the ratio of the number of pixels with slopes within 5°-15° in the i spatial unit to the total number of pixels in the spatial unit; S 0515i represents the number of pixels with slopes within 5°-15° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit. The 15°-35° slope area ratio is the ratio of the number of pixels with a slope within 15°-35° to the total number of pixels in the spatial unit, specifically: IN 1535i =S 1535i / S i Among them: SA 1535i It represents the ratio of the number of pixels with slopes between 15° and 35° in the i spatial unit to the total number of pixels in the spatial unit; S 1535i Indicates the number of pixels with slopes between 15° and 35° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit. The 35°-90° slope area ratio is the ratio of the number of pixels with a slope within 35°-90° to the total number of pixels in the spatial unit, specifically: IN 3590i =S 3590i / S i Among them: SA 3590i It represents the ratio of the number of pixels with slopes between 35° and 90° in the i spatial unit to the total number of pixels in the spatial unit; S 3590i Indicates the number of pixels with slopes between 35° and 90° in the i spatial unit; S i Represents the total number of pixels in the i spatial unit.

5. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The indicators are classified according to the expression of different aspects of terrain, and the terrain fragmentation indicators are summarized into three aspects that can fully express terrain fragmentation.

6. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The entropy weight method calculates the entropy weight of each indicator according to the dispersion degree of each indicator data, and then revises the entropy weight to a certain extent according to each indicator, determines the objective indicator weight, and obtains the comprehensive evaluation value of terrain fragmentation, which is specifically: Positive indicator normalization: Negative indicator normalization: Weight calculation: Among them, m and n represent m samples and n indicators respectively; X ij represents the jth indicator and the specific value of the i-th sample; X 1 ij represents the jth index, the value of the i-th sample after standardization; max(X 1j ,……,X mj ) represents the maximum sample value in the jth index; min(X 1j ,……,X mj ) represents the smallest sample value in the jth index; P ij Represents the weight coefficient of the jth indicator.

7. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The spatial distribution of terrain fragmentation degree is a comprehensive evaluation value of terrain fragmentation based on GIS related software, which uses the natural breakpoint method to visualize the categories to which each spatial unit in the study area belongs.

8. A spatial partitioning method for measuring terrain fragmentation according to claim 1, characterized in that: The spatial filtering is calculated using the values ​​of the spatial unit area.

9. A terrain fragmentation spatial partitioning method according to claim 2, characterized in that: The steps for screening and determining key indicators include: Conduct correlation analysis on the data in the summary table of all selected terrain fragmentation indicators to obtain a correlation matrix diagram; According to the principle of multicollinearity, the absolute values ​​of various indicators in the correlation matrix diagram are compared, and the indicators with absolute values ​​of correlation coefficients higher than a certain threshold (0.75 in this case) are screened, and one of the representative indicators is selected as the key influencing indicator of terrain fragmentation.

10. A terrain fragmentation spatial partitioning method according to claim 1, characterized in that: In the above S6, after obtaining the cluster number corresponding to each spatial unit, the spatial units with the same cluster number are filled with a color, and the structure of the terrain fragmentation spatial partition is displayed in a visualization method of colored blocks.

11. A terrain fragmentation spatial partitioning method, characterized in that: include: A spatial unit acquisition module is used to acquire basic data of terrain fragmentation indicators within a given range, wherein the basic data of terrain fragmentation indicators include DEM grid data of each terrain fragmentation indicator within the county vector range, and the DEM grid data is the basic data for calculating each indicator; Terrain fragmentation index measurement system establishment module: for each terrain fragmentation index, according to the calculation of multiple county unit terrain fragmentation indexes, a terrain fragmentation index summary table is generated; the terrain fragmentation index includes the average slope, elevation variation, and watershed density of the overall terrain fragmentation characteristics, the positive terrain ratio and negative terrain ratio of the positive and negative terrain characteristics, and the slope distribution characteristics of 2°-5°, 5°-15°, 15°-35°, and 35°-90° slope area ratios; Entropy weight evaluation module: used to determine the weight coefficient of terrain fragmentation indicators. First, the data of each indicator is standardized and normalized. After the preliminary data processing, the information entropy and information utility value of each indicator are calculated, and each indicator is normalized. Finally, the weight coefficient of the indicator is calculated. Spatial filtering clustering evaluation module: Based on the idea of ​​spatial filtering, it determines whether the category of each spatial unit is the same as the main category of all adjacent units. If not, the category of the spatial unit is reset to the main category corresponding to its neighboring units. This step can be repeated to obtain a spatial partition result with continuous distribution of the same category and no overlap of different categories.

Images