Three-dimensional green space optimization modeling method, device, equipment and medium based on landscape elevation connectivity
Through the three-dimensional green space optimization modeling method based on landscape elevation connectivity, the problem of difficult to predict the migration path of species along the altitude gradient in traditional technology is solved, and the quantitative evaluation of altitude gradient and the construction of a three-dimensional ecological network are realized, which improves the climate change The accuracy of habitat assessment supports biodiversity conservation.
Patent Information
- Application Number
- CN202411252995.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-09
Smart Images

Figure CN118799499B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of green space modeling, and in particular to a three-dimensional green space optimization modeling method, device, equipment and medium based on landscape elevation connectivity. Background Art
[0002] Currently, climate change has led to accelerated changes in the migration paths and distribution ranges of species. In addition to the empirical conclusion that coordinate shifts occur at the latitude and longitude levels, biological migration also shows a trend of shifting to mountain habitats along the altitude. Green landscapes with high vegetation coverage provide highly suitable habitats for biological communities and can serve as important ecological corridors or stepping stones for species migration to maintain a high number of species. Traditional ecological network simulation technology aims to identify potential ecological corridors based on two-dimensional displacement changes in longitude and latitude. It is still difficult to predict how biological populations migrate along altitude gradients driven by global warming, which greatly limits the effectiveness of biodiversity conservation and restoration.
[0003] However, due to the complex dynamic response of biological individuals or communities to climate change along the altitude gradient, the optimal altitude, the upper limit (front edge) and the lower limit (back edge) of the species do not necessarily move synchronously, or even in the same direction. At present, no indicators have been established to reflect the diffusion restrictions of species across different altitude gradients, and there is still a lack of three-dimensional scene display of anisotropic migration of species in mountainous areas, making it difficult to quantitatively explore the changes in migration paths under the constraints of climate factors. Three-dimensional morphology is actually the addition of altitude or other characteristic information derived from altitude on the basis of two-dimensional pattern. In order to better realize the quantitative characterization of three-dimensional morphology, it is necessary to first construct surface parameters that meet scientific needs, and then adjust the control points to complete the deformation through the idea of "morphing from surface to body". At present, the physical indicators that describe terrain morphology include altitude, slope, slope angle, terrain undulation, etc. These indicators have not yet considered the actual ecological process of species movement and cannot fully reflect the ability of mountain landscape patches to promote or hinder species movement.
[0004] In short, existing technologies mainly focus on the identification of potential ecological corridors in the two-dimensional plane of longitude and latitude, lack the three-dimensional analysis of species migration paths along the altitude driven by future climate change, and largely ignore the highly suitable habitats that can respond to climate change faster and have less resistance. Traditional ecological network modeling technology conducts research based on human activity interference and climate change from the perspective of two-dimensional plane space, and lacks three-dimensional spatial information along the altitude gradient for the study of biological movement ecological processes. The degree to which altitude factors promote / hinder species migration is difficult to describe qualitatively or quantitatively, mainly because the continuous mountain landscape presents a complex fractal structure rather than a monotonically increasing and decreasing vertebral structure. The species migration process does not jump forward at discrete elevation points, but also considers the friction resistance of the entire slope. When the altitude range is within the width of the species niche, the greater the friction resistance, the greater the work done by the species on the slope movement and the more energy consumed, which means that when faced with the need to migrate across altitudes, species tend to choose slopes with small friction resistance, thereby allocating less energy to heat energy and retaining the rest of the energy for various life activities.
[0005] Moreover, the resistance surface assignment is subjective and routine, which is out of touch with the actual migration behavior of species in different mountain habitat patches. It is mainly achieved through source identification, resistance surface superposition, and corridor extraction, focusing on the evaluation of the inhibitory effect of resistance factors such as human activities. The existing simulation methods mainly use expert scoring methods to assign resistance values to the background values of different grid units. This idea assumes that species move on a relatively flat surface and evaluates species accessibility based on the relative barrier capacity of the ground landscape structure. It ignores whether there are diffusion restrictions such as altitude, slope, and slope angle in the actual movement of species across different grid units, resulting in a large deviation in the evaluation of potential adaptive habitats of species driven by future climate change.
[0006] In view of this, this application is filed. Summary of the invention
[0007] The present invention provides a three-dimensional green space optimization modeling method, device, equipment and medium based on landscape elevation connectivity, which can at least partially improve the above-mentioned problems.
[0008] To achieve the above object, the present invention adopts the following technical solutions:
[0009] A three-dimensional green space optimization modeling method based on landscape elevation connectivity includes:
[0010] Obtain the preset minimum hypothesis set principle, construct the metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, where the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid. Each grid unit assumes that the altitude is the only ecological niche feature. When the species moves to different grid units, it corresponds to different altitude suitability;
[0011] Calculate the landscape elevation connectivity results based on the CSV file and the Dijkstra method, and draw the histogram, scatter plot, line graph, and density map of the landscape elevation connectivity results;
[0012] The landscape elevation connectivity results are converted, the layer base height is set, and the LEC boundary section is generated.
[0013] The present invention also provides a three-dimensional green space optimization modeling device based on landscape elevation connectivity, which includes:
[0014] The structure construction unit is used to obtain the preset minimum hypothesis set principle, construct the metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, wherein the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid. Each grid unit assumes that the altitude is the only ecological niche feature. When the species moves to different grid units, it corresponds to different altitude suitability;
[0015] A drawing unit, used to calculate the landscape elevation connectivity results according to the CSV file and the Dijkstra method, and draw a histogram, a scatter plot, a line graph, and a density map of the landscape elevation connectivity results;
[0016] The boundary section generation unit is used to convert the landscape elevation connectivity results, set the layer basic height, and generate the LEC boundary section.
[0017] The present invention also provides a three-dimensional green space optimization modeling device based on landscape elevation connectivity, characterized in that it includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program can be executed by the processor to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of the above items.
[0018] The present invention also provides a computer-readable storage medium, characterized in that it stores a computer program, and the computer program can be executed by a processor of a device where the computer-readable storage medium is located to implement a three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of the above items.
[0019] In summary, the three-dimensional green space optimization modeling method based on landscape elevation connectivity constructs a landscape ecological network based on the altitude gradient, constructs function modules such as plotElevation, init_landscapeConnectivity, computeMinPath, splitRowWise, _test_progress, and computeLEC, and realizes the automatic calculation of landscape elevation connectivity. The method includes initializing landscape elevation connectivity, calculating the minimum path, splitting the calculation domain by row, displaying the progress bar, calculating LEC, writing and viewing the results. Efficient processing is achieved through parallel computing to ensure the accuracy and reliability of the calculation results. The higher the LEC value, the larger the total area of the site and the better the altitude connectivity, and the more likely it is to allow species to "cross mountains and ridges." In addition, this method also customizes parameter optimization of three-dimensional green space modeling, uses ArcScene to display LEC results, and realizes three-dimensional green space landscape optimization and visualization. According to the visual effect of the LEC value, the potential migration path of species in the mountains can be predicted. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a flow chart of a three-dimensional green space optimization modeling method based on landscape elevation connectivity provided by an embodiment of the present invention;
[0021] Figure 2 It is a schematic diagram of the principle of calculating landscape elevation connectivity provided by an embodiment of the present invention;
[0022] Figure 3 is a schematic diagram of the Dijkstra algorithm provided by an embodiment of the present invention, wherein: Figure 3 for Figure 2 The sub-image in the middle frame;
[0023] Figure 4 It is a schematic diagram of splitting the calculation domain by row provided by an embodiment of the present invention;
[0024] Figure 5 This is a comparison diagram of the altitude and LEC results of a certain area in Hainan Tibetan Autonomous Prefecture provided by an embodiment of the present invention;
[0025] Figure 6 A group of graphs of altitude frequency distribution and LEC frequency distribution in a certain area of Hainan Tibetan Autonomous Prefecture provided by an embodiment of the present invention;
[0026] Figure 7 It is a three-dimensional display diagram of the LEC results of a certain area in Hainan Tibetan Autonomous Prefecture provided by an embodiment of the present invention;
[0027] Figure 8 It is a comparison diagram of the altitude and LEC results of a certain area in Huzhou City provided by an embodiment of the present invention;
[0028] Fig. 9It is a group diagram of altitude frequency distribution and LEC frequency distribution in a certain area of Huzhou City provided by an embodiment of the present invention;
[0029] Fig.10 It is a three-dimensional display of LEC results of a certain area in Huzhou City provided by an embodiment of the present invention;
[0030] Fig.11 It is a module schematic diagram of a three-dimensional green space optimization modeling device based on landscape elevation connectivity provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0031] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0032] refer to Figure 1 , Figure 2 As shown, the first embodiment of the present invention discloses a three-dimensional green space optimization modeling method based on landscape elevation connectivity, which can be executed by a three-dimensional green space optimization modeling device based on landscape elevation connectivity (hereinafter referred to as a modeling device), and in particular, executed by one or more processors in the modeling device to implement the following method:
[0033] S1, obtain the preset minimum hypothesis set principle, construct the metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, wherein the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid. Each grid unit assumes that the altitude is the only ecological niche feature. When the species moves to different grid units, it corresponds to different altitude suitability;
[0034] Preferably, the minimum hypothesis set principle includes:
[0035] Individuals of each species have a fitness that depends on altitude, all other vital rates being equal, where the formula for competitive ability is:
[0036]
[0037] in, For individuals at altitude The greatest competitiveness, reflects the competitive ability of individuals of species i at altitude z, is the ecological niche position of species i, is the niche width;
[0038] Different species have different niche positions, but the same niche breadth;
[0039] Niche positions were evenly distributed along the elevation range of the domain, and there was no preferred elevation at the metacommunity scale;
[0040] The dispersion is isotropic;
[0041] The habitat capacity of each local community is constant n.
[0042] Specifically, step S1 includes: obtaining a preset minimum hypothesis set principle, based on the minimum hypothesis set principle, reprojecting the raster file from the latitude and longitude coordinates to the WGS 84_UTM coordinate system, finding the corresponding EPSG code, and specifying the resolution of the elevation grid after projection, to obtain a gridded elevation point matrix, named elev array;
[0043] The constructor plotElevation (data, cmin, cmax, colormap) draws the image of the initial elev array, selects the range of rows and columns from the elev array according to the image viewing effect, deletes the nodata value caused by reprojection, and defines a new elevation array dem based on the clipped elev array, where data is the data set dem to be drawn, cmin and cmax are the range of the color map, and colormap is the color map to be used;
[0044] Specify the X-axis and Y-axis based on the elevation array dem, create X-coordinates and Y-coordinates, use the step parameter to reduce the resolution, and obtain a well-divided equidistant 2D grid;
[0045] Write the divided equidistant 2D grid into a CSV file to obtain the collective community structure.
[0046] In this example, a set of assumptions is specified to specifically explore the effects of landscape structure on species distribution, excluding other possible confounding factors. The assumptions are as follows: (I) Individuals of each species have a fitness that depends on altitude (i.e., competitive ability in this case, formula (1)), and all other vital rates are the same; (II) Different species have different niche positions, but the same niche width; (III) Niche positions are evenly distributed along the elevation range of the domain, so there is no preferred elevation at the metacommunity scale; (IV) Dispersion is isotropic (towards the four nearest local communities in a regular 2D grid); (V) The habitat capacity of each local community is constant at n.
[0047] Therefore, at any time, the metacommunity system is composed of N (local communities) × n (individuals) belonging to S different species. These local communities are organized in an equally spaced 2D grid, and each grid cell assumes altitude as a unique niche feature. When species move to different grid cells, they correspond to different altitude suitability. From the perspective of work and mechanical energy, the absolute value of the difference in suitability between locations at different altitudes is understood as the friction resistance of the slope connecting the two points.
[0048]
[0049] For individuals at altitude The greatest competitiveness, reflects the competitive ability of individuals of species i at altitude z, is the ecological niche position of species i, and the individual is located at altitude z= Its competitiveness is greatest when Represents the niche width and controls the discreteness of the Gaussian function.
[0050] For details, please refer to the following steps:
[0051] Step 1: Reproject the raster grid. A GeoTIFF file is a raster dataset that embeds geographic information (latitude, longitude, map projection, etc.). The geographic data of a GeoTIFF file can be used to position the image at the correct location and geometry to construct structured and unstructured grids used in subsequent landscape evolution models. Reproject the raster file from latitude and longitude coordinates to the WGS 84_UTM (meter) coordinate system and find the corresponding EPSG code at http: / / spatialreference.org / ref / epsg / . Specify the resolution of the elevation grid after projection to obtain a gridded elevation point matrix (essentially a NumPy array containing a 2D slice, named elev).
[0052] Step 2: Crop the elev array. The constructor plotElevation (data, cmin, cmax, colormap) is used to draw the initial elev array image. The range of rows and columns is selected from the elev array according to the viewing effect to remove the nodata value caused by reprojection. A new elevation array dem is defined based on the cropped elev array.
[0053] Step 3: Specify the X and Y axes. In order to maintain the coordinate system for post-processing and possibly reprojecting the output of the LEC in another geospatial system, the X and Y axes need to be specified. The X and Y coordinates are then created, optionally using the step parameter (an integer) to reduce the resolution.
[0054] Step 4: Write the divided equidistant 2D grid into a csv file.
[0055] S2, calculate the landscape elevation connectivity results according to the CSV file and the Dijkstra method, and draw the histogram, scatter plot, line graph, and density map of the landscape elevation connectivity results;
[0056] Specifically, step S2 includes: based on the CSV file, the constructor init_landscapeConnectivity(filename, sigmap=0.1, sigmav=None, connected=True, delimiter=' ', sl=-1.e6,test=False) defines the width of the hypothesized niche and calculates the parameters of the minimum path, where filename (str) is the name of the CSV file containing the regularly spaced elevation grid, sigmap (float) is the species niche width percentage based on altitude, sigmav (float) is the species niche fixed width value, connected (bool) is the path calculated based on diagonal and axial movement, delimiter (str) is the elevation grid CSV delimiter, and sl (float) is the sea level position used to remove ocean points from the LEC calculation;
[0057] According to the parameters of the minimum path, the constructor computeMinPath(r, c) calculates the minimum path between a specific node and all other nodes, creates a cost surface based on the square of the elevation difference between the considered node and all other vertices, creates a "landscape graph" object from the weighted surface, uses the distance-weighted minimum cost path for analysis, calculates the minimum cost distance from the starting cell to all other cells, and generates the landscape elevation connectivity result, where, in the landscape elevation connectivity value, the distance with the lowest cost is calculated with the help of the Dijkstra algorithm in the scikit-image library, r is the row index of the two-dimensional array dem, and c is the column index of the two-dimensional array dem;
[0058] Among them, the formula for the landscape elevation connectivity result is:
[0059]
[0060]
[0061]
[0062]
[0063] in, represents the landscape elevation connectivity at any local community site i, For Site The altitude of is the optimal ecological niche position of the species, is the niche breadth of the species, It measures the closeness of the elevation suitability between site j and site i, where j is any local community site in the N×n metacommunity structure. Measures the proximity of group path j to i along path p, k1, k2, ..., k L are all the community sites included in path p, k1= j, k L = i.
[0064] The constructor splitRowWise() splits the computational domain by rows and uses the message passing interface MPI distributed on multiple processors to calculate the minimum cost path of the points belonging to each subdomain in the entire region using the Dijkstra algorithm;
[0065] Constructor _test_progress(job_title, progress), calculates and displays a progress bar based on the progress parameter to dynamically update progress information, and displays "DONE" when the job is completed, where job_title is the name of the task currently being executed, and progress is a floating point number between 0 and 1, where 0 means the task has just started and 1 means the task has been completed;
[0066] The constructor computeLEC(fout=500) calls the functions computeMinPath(), splitRowWise(), and _test_progress() at the same time to split the array imported from the csv file by row, and parallely calculates the proximity of each node within a similar altitude range to obtain a two-dimensional array representing the landscape height connectivity, and saves the array to the working environment as a csv file;
[0067] The constructor writeLEC(filename='LECout') saves the landscape elevation connectivity results and the original elevation data to disk. In the MPI parallel computing environment, the data is converted from a 2D array to a 1D array, a DataFrame is created and saved as a csv file, and then saved as a vtk file using the gridToVTK function.
[0068] The constructor viewResult(imName=None, size, fsize, cmap1, cmap2, dpi) creates an image window containing two sub-images, displays elevation data on the first sub-image, adds a color bar, displays landscape elevation connectivity data on the second sub-image, adds a color bar, automatically adjusts the sub-image layout, and adds an image title, where size is the size of the drawing window, fsize is the font size, cmap1 is the color map of the first image Elevation, cmap2 is the color map of the second image LEC, and dpi is the image resolution;
[0069] The constructor viewElevFrequency(input=None, imName=None, nbins, size,fsize, dpi) reads data from the specified CSV file, calculates the minimum value of the elevation data, and when it is determined that the minimum value is lower than sea level, uses the sea level value to create a histogram of the elevation data, sets the title and label, and adds a density map of the elevation data on the same map, sets the title and label, where nbins is the number of bins in the histogram, that is, how many intervals the data is divided into;
[0070] The constructor viewLECFrequency(input=None, imName=None, size, fsize, dpi) reads data from the specified CSV file, calculates the minimum value of the elevation data, and when it is determined that the minimum value is lower than sea level, uses the sea level value to create a scatter plot of the LEC data and sets the title and label.
[0071] The constructor viewLECZbar(input=None, imName=None, nbins, size, fsize,dpi) reads data from the specified CSV file, calculates the histogram of elevation data, and calculates the weighted histogram and square weighted histogram based on the LEC data, calculates the mean and standard deviation of each bin, and draws a line graph and a scatter plot of elevation and LEC mean.
[0072] Plot error bars, where represents the standard deviation of each bin, and create a new y-axis that shares the x-axis, plot a density plot of the elevation data, and set the image title, labels, and tick font size.
[0073] In this embodiment, based on the "mid-domain effect" theory, the "species-area" relationship and the classic "island biogeography" theory, three unique geomorphological features of mountain landscapes are identified and analyzed: (a) the limited range of landscape altitude, (b) the frequency distribution of areas at different altitudes, and (c) the altitude connectivity of different locations. Landscape elevation connectivity (LEC) is defined accordingly, which quantifies the proximity of any point in the mountain landscape to all other points at similar altitudes. The LEC value is the sum of the optimal paths from a certain unit in the grid to all other units in the grid. The metric for the optimal path is the altitude suitability of the cell along the species' pathway. The higher the altitude suitability, the greater the probability that the species will pass through the point when choosing a path. This definition can also be seen as a positive treatment of the work done by frictional resistance, that is, the higher the LEC value, the less negative work the species does from the point to all other points in the landscape range, and the lower the energy consumption of the species.
[0074]
[0075]
[0076]
[0077]
[0078] It measures the closeness of the elevation suitability between site j and site i, where j can be any local community site in the N×n metacommunity system. It measures the proximity from group j to i along path p. Assuming that Cji,p is proportional to the product of the colonization probability of each step on path p, the exponential part of the Gaussian fitness function is used to form the formula ,k1,k2,...,k L are all the community sites (k1= j, k L = i).
[0079] Please refer to the following steps for details:
[0080] Step 1: Constructor init_landscapeConnectivity(filename, sigmap=0.1, sigmav=None, connected=True, delimiter=' ', sl=-1.e6, test=False). The meaning of each parameter is as follows:
[0081]
[0082] Step 2: The constructor computeMinPath(r, c) is used to calculate the minimum path between a specific node and all other nodes. A cost surface is created based on the square of the elevation difference between the considered node and all other vertices. A "landscape graph" object is created from the weighted surface, which can then be analyzed using the distance-weighted minimum cost path to calculate the minimum cost distance from the starting cell to all other cells (schematically Figure 3 ). In LEC, we use the Dijkstra algorithm in the scikit-image library to calculate the distance with the lowest cost.
[0083] Step 3: Constructor splitRowWise(). Dijkstra's algorithm is a graph search algorithm that finds the shortest path from a single source on a non-negatively weighted graph. Such an algorithm may take a long time to solve, especially in LEC, because it needs to compute the minimum cost path for each point on the surface. Here we split the computational domain by row and distribute it across multiple processors using the message passing interface (MPI), where the Dijkstra trees for all paths are balanced (see figure). Figure 4 ). Then Dijkstra's algorithm is used in the whole area to calculate the minimum cost path for the points belonging to each subdomain.
[0084] Step 4: Construct the function _test_progress(job_title, progress) to display a visual progress bar of the job progress. The progress bar is calculated and displayed based on the progress parameter, the progress information is dynamically updated, and "DONE" is displayed when the job is completed.
[0085] Step 5: Constructor computeLEC(fout=500). Call the above computeMinPath(), splitRowWise(), and _test_progress() functions to split the array imported from the csv file by row, and parallely calculate the proximity of each node within a similar elevation range. The final result is a two-dimensional array representing the landscape height connectivity, which is saved to the working environment as a csv file.
[0086] Step 6: The constructor writeLEC(filename='LECout') saves the calculated LEC results and the original elevation data to disk. Specifically, it converts the data from a 2D array to a 1D array, creates a DataFrame and saves it as a csv file, and uses the gridToVTK function to save it as a vtk file. The entire process is performed in an MPI parallel computing environment.
[0087] Step 7: Constructor viewResult(imName=None, size, fsize, cmap1, cmap2,dpi). Create an image window with two sub-images, display elevation data on the first sub-image, add a color bar, display landscape elevation connectivity data on the second sub-image, add a color bar. Automatically adjust the sub-image layout and add an image title.
[0088] Step 8: Constructor viewElevFrequency(input=None, imName=None, nbins, size,fsize, dpi). Read data from the specified CSV file. Calculate the minimum value of the elevation data, and if the value is below sea level, use the sea level value. Create a histogram of the elevation data and set the title and labels. Add a density plot of the elevation data on the same plot and set the title and labels.
[0089] Step 9: Constructor viewLECFrequency(input=None, imName=None, size, fsize,dpi). Read data from the specified CSV file. Calculate the minimum value of the elevation data. If the value is below sea level, use the sea level value. Create a scatter plot of the LEC data and set the title and labels.
[0090] Step 10: Constructor viewLECZbar(input=None, imName=None, nbins, size,fsize, dpi). Read data from the specified CSV file, calculate the histogram of the elevation data, and calculate the weighted histogram and squared weighted histogram based on the LEC data. Calculate the mean and standard deviation for each bin. Plot a line and scatter plot of elevation versus LEC mean. Plot error bars representing the standard deviation for each bin. Create a new y-axis that shares the x-axis and plot a density plot of the elevation data. Finally, set the image title, label, and tick font size.
[0091] S3, convert the landscape elevation connectivity results, set the layer base height, and generate the LEC boundary section.
[0092] Specifically, step S3 includes: using the arcpy library to convert the landscape elevation connectivity result into a GeoTiff raster file, wherein arcpy.CreateFeatureclass_management is used to create a point feature class, arcpy.AddField_management is used to add fields, and arcpy.InsertCursor is used to insert data row by row with a cursor; arcpy.conversion.PointToRaster is used to convert the point vector file into a raster file LEC.tif;
[0093] Use ArcScene to set the layer base height and generate the LEC boundary section.
[0094] In this embodiment, the three-dimensional green landscape is optimized by custom parameters; first, the arcpy library is used to convert the LEC calculation results into GeoTiff raster files. Specifically, a point feature class is created through arcpy.CreateFeatureclass_management, fields are added through arcpy.AddField_management, data is inserted row by row with a cursor through arcpy.InsertCursor, and finally the point vector file is converted into a raster file LEC.tif through arcpy.conversion.PointToRaster. Secondly, ArcScene is used to set the base height of the layer. Import the landscape elevation connectivity result LEC.tif and the elevation file dem.tif, right-click the LEC.tif properties, and select Float on the custom surface dem.tif under the Base Heights column. Click the raster resolution, which is recommended to be consistent with the equidistant 2D grid divided by LEC. Set the unit coefficient. The larger the value, the more obvious the terrain changes (it is recommended to be set to 6). The image effect can also be adjusted by setting the rendering properties of LEC.tif, and select it in combination with computer conditions. Finally, the LEC boundary section is generated. Use the [3D Analyst Tools] - [Conversion] - [FromRaster] - [Raster Domain] tool, input the raster LEC.tif, select the Line type for the output feature class type, and output as domain. Right-click the domain property, select Float on the custom surface dem.tif under the Base Heights column, and set the unit factor to 6. Check "Extrude features in layer" under the Extrusion column, set the extrusion value to -250 (negative numbers mean stretching downward), and the extrusion method to "Use this as the value to which the feature is extruded".
[0095] See also Figures 5 to 7In this embodiment, Hainan Tibetan Autonomous Prefecture is used as an example for explanation:
[0096] Step 1: Use the advanced search method to download the 30mDEM digital elevation data product (https: / / www.gscloud.cn / home) according to the administrative region "Qinghai Province-Hainan Tibetan Autonomous Prefecture-Xinghai County". In this example, the ASTGTMV003_N35E099 video is downloaded.
[0097] Step 2: Import the ASTGTMV003_N35E099_dem.tif layer in ArcMap, create a new polygon shapefile, select the area with obvious elevation fluctuations, draw a rectangle and mask it. Combined with the computer memory load status, in order to control the number of rows and columns of the subsequent calculation grid to about 1000, resample ASTGTMV003_N35E099_dem.tif to cellsize (X,Y) equal to (0.00083, 0.00083), and finally export it as rec_dem.tif.
[0098] Step 3: Reproject the raster grid. Reproject the raster file from latitude and longitude coordinates to the WGS 84_UTM_zone_47N coordinate system, corresponding to the EPSG code 32647. Specify the resolution of the elevation grid after projection as 100 m, and obtain the gridded elevation point matrix (elev array).
[0099] Step 4: Crop the elev array. Plot the elev array with plotElevation (elev, 2500, 5500, topocmap), select the range of rows and columns to be cropped elev[0:500,0:800] according to the view effect and plot again. Check whether the nodata value is completely deleted with np.ma.is_masked (elev[0:500,0:800]). Define a new elevation array dem based on the cropped elev array.
[0100] Step 5: Specify the X and Y axes, create the X and Y coordinates, and write the divided equidistant 2D grid (dem array) to the csv file hnrec_100.csv.
[0101] Step 6: Define the file path of the elevation data, call the function init_landscapeConnectivity, pass in the elevation data file path, and initialize the variables and data required for calculation. The landscape elevation connectivity of the area is calculated by computeLEC.
[0102] Step 7: Write the results. Call the writeLEC function to write the calculated LEC results to disk, save them as result_hainan.csv file, and generate a VTK visualization file.
[0103] Step 8: View the results. Call the viewResult function to generate and save an image containing elevation data and LEC data as plot_hainan.png. Call the viewElevFrequency function to generate and save a frequency histogram and density map of elevation data as elev_freq_hainan.png with a resolution of 300dpi. Call the viewLECFrequency function to generate and save a scatter plot of LEC versus elevation as lec_freq_hainan.png with a resolution of 300dpi. Call the viewLECZbar function to generate and save an error bar plot and density map of LEC versus elevation as lec_bar_hainan.png with a resolution of 300dpi.
[0104] Step 9: Use the arcpy library to convert the LEC calculation results (result_hainan.csv) into a GeoTiff raster file (hainan_LEC.tif).
[0105] Step 10: Use ArcScene to set the floating height of the layer. Import the landscape elevation connectivity result hainan_LEC.tif and the elevation file ASTGTMV003_N35E099_dem.tif. First, use the [Data ManagementTools]-[Projections and Transformations]-[Raster]-[Project Raster] tool to reproject the layer ASTGTMV003_N35E099_dem.tif to WGS_1984_UTM_Zone_47N and save it as N35E099_demUTM47.tif. Second, right-click on the hainan_LEC.tif properties, select Floating on the custom surface N35E099_demUTM47.tif under the Base Heights column, and click the raster resolution to set it to 100 m (consistent with the equidistant 2D grid divided by LEC). Set the unit coefficient value to 6. Adjust the slider under the Rendering column to maximize the raster image quality and minimize the minimum transparency threshold. Finally, right-click on the N35E099_demUTM47.tif properties, select Float on Custom Surface N35E099_demUTM47.tif under the Base Heights column, click on the grid resolution and set it to 30 m. Set the Unit Factor value to 3. Under the Rendering column, adjust the slider to make the raster image quality the highest and the minimum transparency threshold the lowest.
[0106] Step 11: Generate the LEC boundary section. Use the [3D Analyst Tools] — [Conversion] — [FromRaster] — [Raster Domain] tool, input the raster hainan_LEC.tif, select the Line type for the output feature class type, and output as domain. Right-click the domain property, select Float on the custom surface N35E099_demUTM47.tif under the Base Heights column, and set the unit factor to 6. Check "Extrude features in layer" under the Extrusion column, set the extrusion value to -250 (negative numbers mean stretching downward), and the extrusion method to "Use this as the value to which the feature is extruded".
[0107] See also Figures 7 to 10 In this embodiment, Huzhou City is used as an example to illustrate again:
[0108] Step 1: Use the advanced search method to download the 30m DEM digital elevation data product (https: / / www.gscloud.cn / home) according to the administrative region "Zhejiang Province-Huzhou City-Anji County". In this example, the ASTGTMV003_N30E119 video is downloaded.
[0109] Step 2: Import the ASTGTMV003_N30E119_dem.tif layer in ArcMap, create a new polygon shapefile, select the area with obvious elevation fluctuations, draw a rectangle and mask it. Combined with the computer memory load status, in order to control the number of rows and columns of the subsequent calculation grid to about 1000, resample ASTGTMV003_N30E119_dem.tif to cellsize (X,Y) equal to (0.00083, 0.00083), and finally export it as rec_dem.tif.
[0110] Step 3: Reproject the raster grid. Reproject the raster file from latitude and longitude coordinates to the WGS 84_UTM_zone_50N coordinate system, corresponding to the EPSG code 32650. Specify the resolution of the elevation grid after projection as 100 m, and obtain the gridded elevation point matrix (elev array).
[0111] Step 4: Crop the elev array. Plot the elev array with plotElevation (elev, 0, 1600, topocmap), select the range of rows and columns to be cropped elev[20:800, 50:800] according to the view effect and plot again. Check whether the nodata value is completely deleted with np.ma.is_masked (elev[20:800, 50:800]). Define a new elevation array dem based on the cropped elev array.
[0112] Step 5: Specify the X and Y axes, create X and Y coordinates, and write the divided equidistant 2D grid (dem array) to the csv file hzrec_100.csv.
[0113] Step 6: Define the file path of the elevation data, call the function init_landscapeConnectivity, pass in the elevation data file path, and initialize the variables and data required for calculation. The landscape elevation connectivity of the area is calculated by computeLEC.
[0114] Step 7: Write the results. Call the writeLEC function to write the calculated LEC results to disk, save them as result_huzhou.csv file, and generate a VTK visualization file.
[0115] Step 8: View the results. Call the viewResult function to generate and save an image containing elevation data and LEC data as plot_huzhou.png file. Call the viewElevFrequency function to generate and save the frequency histogram and density map of elevation data as elev_freq_huzhou.png file with a resolution of 300dpi. Call the viewLECFrequency function to generate and save a scatter plot of LEC versus elevation as lec_freq_huzhou.png file with a resolution of 300dpi. Call the viewLECZbar function to generate and save an error bar plot and density map of LEC versus elevation as lec_bar_huzhou.png file with a resolution of 300dpi.
[0116] Step 9: Use the arcpy library to convert the LEC calculation results (result_huzhou.csv) into a GeoTiff raster file (huzhou_LEC.tif).
[0117] Step 10: Use ArcScene to set the floating height of the layer. Import the landscape elevation connectivity result huzhou_LEC.tif and the elevation file ASTGTMV003_N30E119_dem.tif. First, use the [Data ManagementTools]-[Projections and Transformations]-[Raster]-[Project Raster] tool to reproject the layer ASTGTMV003_N30E119_dem.tif to WGS_1984_UTM_Zone_50N and save it as N30E119_demUTM50.tif. Second, right-click the huzhou_LEC.tif properties, select Floating on the custom surface N30E119_demUTM50.tif under the Base Heights column, and click the raster resolution to set it to 100 m (consistent with the equidistant 2D grid divided by LEC). Set the unit coefficient value to 6. Adjust the slider under the Rendering column to maximize the raster image quality and minimize the minimum transparency threshold. Finally, right-click on the N30E119_demUTM50.tif properties, select Float on Custom Surface N30E119_demUTM50.tif under the Base Heights column, click on the grid resolution and set it to 30 m. Set the Unit Factor value to 3. Under the Rendering column, adjust the slider to make the raster image quality the highest and the minimum transparency threshold the lowest.
[0118] Step 11: Generate LEC boundary section. Use [3D Analyst Tools] — [Conversion] — [FromRaster] — [Raster Domain] tool, input raster huzhou_LEC.tif, select Line type for output feature class type, and output as domain. Right-click domain attribute, select Float on custom surface N30E119_demUTM50.tif under Base Heights column, and set unit coefficient to 6. Check "Extrude features in layer" under Extrusion column, set extrusion value to -250 (negative number means downward stretching), and set extrusion method to "Use it as the value to which the feature is extruded".
[0119] The three-dimensional green space optimization modeling method based on landscape elevation connectivity aims to integrate the altitude changes of mountain landscapes, fully reflect the potential corridors for species habitat migration along altitude gradients driven by future climate change, provide a three-dimensional green space optimization modeling method based on landscape elevation connectivity, and use code to write a module that can realize full-process automation. Drawing on the core connotation of the ecological network, the main solutions are: 1. Breaking the traditional landscape connectivity modeling technology routine, considering the actual movement process of species, and quantifying the ability of altitude gradients to promote or hinder species migration. 2. Customized parameter optimization three-dimensional modeling, parameters can scientifically characterize the characteristics of regional topography and geomorphology.
[0120] In summary, there is an urgent need to achieve three-dimensional spatial optimization of species migration paths along altitude gradients through three-dimensional reconstruction technology, so as to provide technical support for the study of adaptive strategies of mountain species to cope with climate change. Ideally, the optimization of three-dimensional mountain ecological networks based on altitude gradients will reduce the risk of species exposure to macroclimate warming and extreme high temperature events. This method quantitatively evaluates the promotion-inhibition coupling effect of altitude gradients on species migration, constructs a high-resolution three-dimensional green space landscape optimization model, reveals the adaptive three-dimensional ecological network of biodiversity distribution pattern, and analyzes the priority protection areas of mountain landscapes driven by climate change, while fully considering the diversity and geographic spatial complexity of mountain ecosystems.
[0121] See also Fig.11 The second embodiment of the present invention further provides a three-dimensional green space optimization modeling device based on landscape elevation connectivity, which includes:
[0122] The structure construction unit 201 is used to obtain a preset minimum hypothesis set principle, construct a metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, wherein the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid, and each grid unit assumes that the altitude is a unique niche feature. When the species moves to different grid units, it corresponds to different altitude suitability;
[0123] A drawing unit 202 is used to calculate the landscape elevation connectivity result according to the CSV file and the Dijkstra method, and draw a histogram, a scatter plot, a line graph, and a density map of the landscape elevation connectivity result;
[0124] The boundary section generating unit 203 is used to convert the landscape elevation connectivity result, set the layer basic height, and generate the LEC boundary section;
[0125] The minimum hypothesis set principle includes:
[0126] Individuals of each species have a fitness that depends on altitude, all other vital rates being equal, where the formula for competitive ability is:
[0127]
[0128] in, For individuals at altitude The greatest competitiveness, reflects the competitive ability of individuals of species i at altitude z, is the ecological niche position of species i, is the niche width;
[0129] Different species have different niche positions, but the same niche breadth;
[0130] Niche positions were evenly distributed along the elevation range of the domain, and there was no preferred elevation at the metacommunity scale;
[0131] The dispersion is isotropic;
[0132] The habitat capacity of each local community is constant, n;
[0133] The landscape elevation connectivity results are calculated based on the CSV file and the Dijkstra method, as follows:
[0134] Based on the CSV file, the constructor init_landscapeConnectivity(filename, sigmap=0.1,sigmav=None, connected=True, delimiter=' ', sl=-1.e6, test=False) defines the width of the hypothesized niche and calculates the parameters of the minimum path, where filename (str) is the name of the CSV file containing the regularly spaced elevation grid, sigmap (float) is the percentage of species niche width based on altitude, sigmav (float) is the fixed width value of the species niche, connected (bool) is the path calculated based on diagonal and axial movement, delimiter (str) is the elevation grid CSV delimiter, and sl (float) is the sea level position used to remove ocean points from the LEC calculation;
[0135] According to the parameters of the minimum path, the constructor computeMinPath(r, c) calculates the minimum path between a specific node and all other nodes, creates a cost surface based on the square of the elevation difference between the considered node and all other vertices, creates a "landscape graph" object from the weighted surface, uses the distance-weighted minimum cost path for analysis, calculates the minimum cost distance from the starting cell to all other cells, and generates the landscape elevation connectivity result, where, in the landscape elevation connectivity value, the distance with the lowest cost is calculated with the help of the Dijkstra algorithm in the scikit-image library, r is the row index of the two-dimensional array dem, and c is the column index of the two-dimensional array dem;
[0136] Among them, the formula for the landscape elevation connectivity result is:
[0137]
[0138]
[0139]
[0140]
[0141] in, represents the landscape elevation connectivity at any local community site i, For Site The altitude of is the optimal ecological niche position of the species, is the niche breadth of the species, It measures the closeness of the elevation suitability between site j and site i, where j can be any local community site in the N×n metacommunity structure. Measures the proximity of group path j to i along path p, k1, k2, ..., k L are all the community sites included in path p, k1= j, k L = i.
[0142] The third embodiment of the present invention also provides a three-dimensional green space optimization modeling device based on landscape elevation connectivity, characterized in that it includes a memory and a processor, the memory stores a computer program, and the computer program can be executed by the processor to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of the above items.
[0143] The fourth embodiment of the present invention also provides a computer-readable storage medium, characterized in that a computer program is stored therein, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of the above items.
[0144] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A three-dimensional green space optimization modeling method based on landscape elevation connectivity, characterized in that: include: Obtain the preset minimum hypothesis set principle, construct the metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, where the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid. Each grid unit assumes that the altitude is the only ecological niche feature. When the species moves to different grid units, it corresponds to different altitude suitability; Calculate the landscape elevation connectivity results based on the CSV file and the Dijkstra method, and draw the histogram, scatter plot, line graph, and density map of the landscape elevation connectivity results; The landscape elevation connectivity results are converted, the base height of the layer is set, and the LEC boundary section is generated; The minimum hypothesis set principle includes: Individuals of each species have a fitness that depends on altitude, all other vital rates being equal, where the formula for competitive ability is: in, For individuals at altitude The greatest competitiveness, reflects the competitive ability of individuals of species i at altitude z, is the ecological niche position of species i, is the niche width; Different species have different niche positions, but the same niche breadth; Niche positions were evenly distributed along the elevation range of the domain, and there was no preferred elevation at the metacommunity scale; The dispersion is isotropic; The habitat capacity of each local community is constant, n; The landscape elevation connectivity results are calculated based on the CSV file and the Dijkstra method, as follows: Based on the CSV file, the constructor init_landscapeConnectivity(filename, sigmap=0.1,sigmav=None, connected=True, delimiter=' ', sl=-1.e6, test=False) defines the width of the hypothesized niche and calculates the parameters of the minimum path, where filename (str) is the name of the CSV file containing the regularly spaced elevation grid, sigmap (float) is the percentage of species niche width based on altitude, sigmav (float) is the fixed width value of the species niche, connected (bool) is the path calculated based on diagonal and axial movement, delimiter (str) is the elevation grid CSV delimiter, and sl (float) is the sea level position used to remove ocean points from the LEC calculation; According to the parameters of the minimum path, the constructor computeMinPath(r, c) calculates the minimum path between a specific node and all other nodes, creates a cost surface based on the square of the elevation difference between the considered node and all other vertices, creates a "landscape graph" object from the weighted surface, uses the distance-weighted minimum cost path for analysis, calculates the minimum cost distance from the starting cell to all other cells, and generates the landscape elevation connectivity result, where the Dijkstra algorithm in the scikit-image library is used to calculate the distance with the lowest cost in the landscape elevation connectivity value, r is the row index of the two-dimensional array dem, and c is the column index of the two-dimensional array dem; Among them, the formula for the landscape elevation connectivity result is: in, represents the landscape elevation connectivity at any local community site i, For Site The altitude of is the optimal ecological niche position of the species, is the niche breadth of the species, It measures the closeness of the elevation suitability between site j and site i, where j is any local community site in the N×n metacommunity structure. Measures the proximity of group path j to i along path p, k1, k2, ..., k L are all the community sites included in path p, k1 = j, k L = i.
2. The three-dimensional green space optimization modeling method based on landscape elevation connectivity according to claim 1 is characterized in that: Obtain the preset minimum hypothesis set principle, construct the collective community structure according to the minimum hypothesis set principle, and generate a CSV file, specifically: Get the preset minimum hypothesis set principle, based on the minimum hypothesis set principle, reproject the raster file from the latitude and longitude coordinates to the WGS 84_UTM coordinate system, find the corresponding EPSG code, and specify the resolution of the elevation grid after projection to obtain the gridded elevation point matrix, named elev array; The constructor plotElevation (data, cmin, cmax, colormap) draws the image of the initial elev array, selects the range of rows and columns from the elev array according to the image viewing effect, deletes the nodata value caused by reprojection, and defines a new elevation array dem based on the clipped elev array, where data is the data set dem to be drawn, cmin and cmax are the range of the color map, and colormap is the color map to be used; Specify the X-axis and Y-axis based on the elevation array dem, create X-coordinates and Y-coordinates, use the step parameter to reduce the resolution, and obtain a well-divided equidistant 2D grid; Write the divided equidistant 2D grid into a CSV file to obtain the collective community structure.
3. The three-dimensional green space optimization modeling method based on landscape elevation connectivity according to claim 1 is characterized in that: Draw histograms, scatter plots, line graphs, and density maps of landscape elevation connectivity results: The constructor splitRowWise() splits the computational domain by rows and uses the message passing interface MPI distributed on multiple processors to calculate the minimum cost path of the points belonging to each subdomain in the entire region using the Dijkstra algorithm; Constructor _test_progress(job_title, progress), calculates and displays a progress bar based on the progress parameter to dynamically update progress information, and displays "DONE" when the job is completed, where job_title is the name of the task currently being executed, and progress is a floating point number between 0 and 1, where 0 means the task has just started and 1 means the task has been completed; The constructor computeLEC(fout=500) calls the functions computeMinPath(), splitRowWise(), and _test_progress() at the same time to split the array imported from the csv file by row, and parallely calculates the proximity of each node within a similar altitude range to obtain a two-dimensional array representing the landscape height connectivity, and saves the array to the working environment as a csv file; The constructor writeLEC(filename='LECout') saves the landscape elevation connectivity results and the original elevation data to disk. In the MPI parallel computing environment, the data is converted from a 2D array to a 1D array, a DataFrame is created and saved as a csv file, and then saved as a vtk file using the gridToVTK function. The constructor viewResult(imName=None, size, fsize, cmap1, cmap2, dpi) creates an image window containing two sub-images, displays elevation data on the first sub-image, adds a color bar, displays landscape elevation connectivity data on the second sub-image, adds a color bar, automatically adjusts the sub-image layout, and adds an image title, where size is the size of the drawing window, fsize is the font size, cmap1 is the color map of the first image Elevation, cmap2 is the color map of the second image LEC, and dpi is the image resolution; The constructor viewElevFrequency(input=None, imName=None, nbins, size, fsize,dpi) reads data from the specified CSV file, calculates the minimum value of the elevation data, and when it is determined that the minimum value is lower than sea level, uses the sea level value to create a histogram of the elevation data, sets the title and label, and adds a density map of the elevation data on the same map, sets the title and label, where nbins is the number of bins in the histogram, that is, how many intervals the data is divided into; The constructor viewLECFrequency(input=None, imName=None, size, fsize, dpi) reads data from the specified CSV file, calculates the minimum value of the elevation data, and when it is determined that the minimum value is lower than sea level, uses the sea level value to create a scatter plot of the LEC data and sets the title and label. The constructor viewLECZbar(input=None, imName=None, nbins, size, fsize, dpi) reads data from the specified CSV file, calculates the histogram of the elevation data, and calculates the weighted histogram and square weighted histogram based on the LEC data, calculates the mean and standard deviation of each bin, and draws a line chart and a scatter plot of the elevation and the LEC mean; draws error bars to represent the standard deviation of each bin, and creates a new y-axis that shares the x-axis, draws a density map of the elevation data, and sets the image title, label, and scale font size.
4. The three-dimensional green space optimization modeling method based on landscape elevation connectivity according to claim 1 is characterized in that: The landscape elevation connectivity results are converted, the basic height of the layer is set, and the LEC boundary section is generated, specifically: The arcpy library is used to convert the landscape elevation connectivity results into GeoTiff raster files. Arcpy.CreateFeatureclass_management is used to create a point feature class, arcpy.AddField_management is used to add fields, and arcpy.InsertCursor is used to insert data row by row with a cursor. The point vector file is converted into a raster file LEC.tif by using arcpy.conversion.PointToRaster. Use ArcScene to set the layer base height and generate the LEC boundary section.
5. A three-dimensional green space optimization modeling device based on landscape elevation connectivity, used to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as claimed in claim 1, characterized in that: include: The structure construction unit is used to obtain the preset minimum hypothesis set principle, construct the metacommunity structure according to the minimum hypothesis set principle, and generate a CSV file, wherein the metacommunity structure is composed of local community organizations of N local communities × n individuals belonging to S different species, and the local community organizations are organized in an equally spaced 2D grid. Each grid unit assumes that the altitude is the only ecological niche feature. When the species moves to different grid units, it corresponds to different altitude suitability; A drawing unit, used to calculate the landscape elevation connectivity results according to the CSV file and the Dijkstra method, and draw a histogram, a scatter plot, a line graph, and a density map of the landscape elevation connectivity results; The boundary section generation unit is used to convert the landscape elevation connectivity results, set the layer base height, and generate the LEC boundary section; The minimum hypothesis set principle includes: Individuals of each species have a fitness that depends on altitude, all other vital rates being equal, where the formula for competitive ability is: in, For individuals at altitude The greatest competitiveness, reflects the competitive ability of individuals of species i at altitude z, is the ecological niche position of species i, is the niche width; Different species have different niche positions, but the same niche breadth; Niche positions were evenly distributed along the elevation range of the domain, and there was no preferred elevation at the metacommunity scale; The dispersion is isotropic; The habitat capacity of each local community is constant, n; The landscape elevation connectivity results are calculated based on the CSV file and the Dijkstra method, as follows: Based on the CSV file, the constructor init_landscapeConnectivity(filename, sigmap=0.1,sigmav=None, connected=True, delimiter=' ', sl=-1.e6, test=False) defines the width of the hypothesized niche and calculates the parameters of the minimum path, where filename (str) is the name of the CSV file containing the regularly spaced elevation grid, sigmap (float) is the percentage of species niche width based on altitude, sigmav (float) is the fixed width value of the species niche, connected (bool) is the path calculated based on diagonal and axial movement, delimiter (str) is the elevation grid CSV delimiter, and sl (float) is the sea level position used to remove ocean points from the LEC calculation; According to the parameters of the minimum path, the constructor computeMinPath(r, c) calculates the minimum path between a specific node and all other nodes, creates a cost surface based on the square of the elevation difference between the considered node and all other vertices, creates a "landscape graph" object from the weighted surface, uses the distance-weighted minimum cost path for analysis, calculates the minimum cost distance from the starting cell to all other cells, and generates the landscape elevation connectivity result, where the Dijkstra algorithm in the scikit-image library is used to calculate the distance with the lowest cost in the landscape elevation connectivity value, r is the row index of the two-dimensional array dem, and c is the column index of the two-dimensional array dem; Among them, the formula for the landscape elevation connectivity result is: in, represents the landscape elevation connectivity at any local community site i, For Site The altitude of is the optimal ecological niche position of the species, is the niche breadth of the species, It measures the closeness of the elevation suitability between site j and site i, where j is any local community site in the N×n metacommunity structure. Measures the proximity of group path j to i along path p, k1, k2, ..., k L are all the community sites included in path p, k1 = j, k L = i.
6. A three-dimensional green space optimization modeling device based on landscape elevation connectivity, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of claims 1 to 4.
7. A computer-readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement the three-dimensional green space optimization modeling method based on landscape elevation connectivity as described in any one of claims 1 to 4.
