A method and system for establishing a mountainous tree species afforestation area containing bamboo species

Abstract

The present application provides a kind of mountainous tree species afforestation area establishment method and system containing bamboo species, it is related to afforestation scientific and technical field.The present application obtains the mountainous tree species distribution point data carrying geographic coordinates;Obtain the basin boundary data, if it cannot be obtained, obtain the digital elevation data covering the study area and demarcate the basin boundary accordingly;According to the distribution point in the basin boundary, select and combine one or more basins to obtain the mountainous basin range;Obtain the range digital elevation data and generate terrain spatial attribute raster data, the spatial attribute includes elevation, potential solar radiation, slope, cosine / sine conversion slope direction, convergence area, shadow index, convergence intensity index and terrain humidity index;According to the distribution point coordinate, the corresponding spatial attribute value is extracted to form a training sample set;Based on the sample set, a model is established and evaluated, and the suitable afforestation area is output by predicting the basin range. Terrain attribute is used instead of climate and soil interpolation data, which is suitable for stable prediction under sparse monitoring point conditions.

Description

Technical Field

[0001] This invention relates to the field of afforestation science and technology, and in particular to a method and system for establishing afforestation areas in mountainous areas containing bamboo species. Background Technology

[0002] With population growth and economic development, most of the Earth's native forests have been destroyed, especially in plains areas. Therefore, mountainous regions have become the main areas of forest distribution, and are also biodiversity hotspots, serving as refuges for biodiversity and less affected by human activities. Afforestation is a nature-based solution to address environmental degradation, climate change, and biodiversity loss.

[0003] Selecting suitable afforestation sites is a prerequisite for successful afforestation activities in mountainous areas. Traditionally, afforestation areas are typically established using species distribution models, which link known species distributions with the spatial distribution of environmental variables to create models and predict suitable afforestation areas. These models usually select spatial raster data on climate and soil as predictor variables. However, these data, especially climate and soil data, are highly uncertain because the complex topography of mountainous areas leads to significant climate and soil variations over short distances, making it difficult to create accurate climate and soil spatial raster data. Furthermore, mountainous areas typically have fewer meteorological stations and soil sampling points, and using these point data for spatial interpolation to obtain areal data also introduces significant errors. More importantly, the data obtained through spatial interpolation cannot reflect the micro-environmental characteristics of small topographical areas and microhabitats. In summary, mountainous areas lack high-resolution spatial raster data that reflects the micro-habitat characteristics of species. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for establishing afforestation areas for mountainous tree species containing bamboo. By using topographic spatial attributes extracted from digital elevation data to replace traditional climate and soil interpolation data, stable prediction of suitable afforestation areas for mountainous tree species is achieved under conditions of sparse monitoring points.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] A method for establishing afforestation areas for tree species in mountainous regions containing bamboo includes:

[0007] Acquire data on the distribution points of tree species in mountainous areas; the data on the distribution points of tree species in mountainous areas carries geographical coordinates.

[0008] Obtain watershed boundary data; if the watershed boundary data cannot be obtained, obtain digital elevation data covering the area where the distribution point data of the mountain tree species is located, and delineate the watershed boundary based on the digital elevation data to form the watershed boundary data.

[0009] Based on the distribution data of tree species in the mountainous area, one or more watersheds are selected from the watershed boundary data and combined to obtain the watershed range of the mountainous area;

[0010] Topographic map data of the mountainous watershed is obtained based on the watershed area of ​​the mountainous area; the topographic map data of the mountainous watershed is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal;

[0011] Spatial attribute raster data is extracted based on the digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index;

[0012] Based on the geographic coordinates of the mountain tree species distribution point data, corresponding spatial attribute values ​​are extracted from the spatial attribute raster data to form a training sample set with the mountain tree species distribution point data as the response variable and the spatial attribute values ​​as the prediction variable.

[0013] A model is established based on the training sample set, and the predictive performance of the model is evaluated. After the model passes the evaluation, the spatial attribute raster data is used to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species. The model can be either a statistical model or an artificial intelligence model.

[0014] Preferably, acquiring data on the distribution points of tree species in mountainous areas includes:

[0015] Data on the distribution points of tree species in mountainous areas with geographic coordinates are obtained, and the location accuracy of the geographic coordinates is at least 30 meters; the data sources for the distribution points of tree species in mountainous areas include at least one or more of the following: field survey data, forest resource inventory data or forest resource data released by forest resource management departments, vegetation maps or remote sensing map data, published literature data, and online database data.

[0016] Preferably, acquiring watershed boundary data includes:

[0017] Obtain free online shared watershed boundary data, or digitize a paper map of the watershed distribution in the region to obtain the watershed boundary data; wherein, the free online shared watershed boundary data includes at least the watershed boundary data shared by the global watershed boundary data HydroBASINS or the National Basic Science Data Center.

[0018] Preferably, when the watershed boundary data cannot be obtained, digital elevation data covering the area where the distribution data of the mountain tree species is located is obtained, and the watershed boundary is delineated based on the digital elevation data to form the watershed boundary data, including:

[0019] The watershed extent is extracted from the digital elevation data using the hydrological analysis module of the geographic information system software to form the watershed boundary data; wherein, the digital elevation data is the digital elevation data after vegetation and building height correction and removal.

[0020] Preferably, based on the distribution point data of tree species in the mountainous area, one or more watersheds are selected from the watershed boundary data and combined to obtain the watershed range of the mountainous area, including:

[0021] One or more watersheds are selected based on the geographic coordinates of the mountain tree species distribution point data, such that each of the mountain tree species distribution point data falls within the combined boundary of the one or more watersheds.

[0022] The combined boundary range of the one or more watersheds is defined as the range of the mountainous watershed, and the total area of ​​the watershed corresponding to the range of the mountainous watershed is controlled within 1,000 square kilometers.

[0023] Preferably, obtaining topographic map data of the mountainous watershed based on the watershed area includes:

[0024] Obtain digital elevation data as the topographic map data of the mountainous watershed, wherein the digital elevation data is obtained in any of the following ways:

[0025] Perform georegistration on paper topographic maps to achieve digitization, and obtain the digital elevation data through the terrain-to-raster conversion function; or

[0026] Apply for topographic map data from national and local natural resources management departments at all levels, and obtain the digital elevation data from the topographic map data; or

[0027] Download the digital elevation data after vegetation and building height correction and removal, and use the corrected and removed digital elevation data as the topographic map data of the mountainous watershed.

[0028] Preferably, extracting spatial attribute raster data based on the digital elevation data includes:

[0029] The slope is extracted from the digital elevation data using the slope function;

[0030] The slope aspect raster data is extracted from the digital elevation data using the slope aspect function, and cosine and sine function operations are performed on the slope aspect raster data respectively to obtain the cosine-converted slope aspect and the sine-converted slope aspect;

[0031] The potential solar radiation is extracted from the digital elevation data using the solar radiation point function;

[0032] The shadow index is extracted from the digital elevation data using the mountain shadow function;

[0033] Accumulated flow raster data is extracted from the digital elevation data using the flow accumulation function to generate the runoff area, the runoff intensity index, and the topographic wetness index.

[0034] Preferably, based on the geographic coordinates of the distribution points of tree species in the mountainous area, corresponding spatial attribute values ​​are extracted from the spatial attribute raster data to form a training sample set with the distribution points of tree species in the mountainous area as the response variable and the spatial attribute values ​​as the predictor variable, including:

[0035] The distribution data of tree species in the mountainous area is used as the response variable, and the response variable is set as a binary variable, where 1 indicates that the tree species exists and 0 indicates that the tree species does not exist.

[0036] Using the "Extract Values ​​to Points" function module in Geographic Information System software or R language software, the corresponding spatial attribute values ​​are extracted from the spatial attribute raster data according to the geographic coordinates of the distribution points of tree species in the mountainous area, and used as the prediction variables;

[0037] The response variable is paired with the predictor variable to form the training sample set.

[0038] Preferably, a model is built based on the training sample set, and the predictive performance of the model is evaluated. After the model passes the evaluation, the spatial attribute raster data is used to predict the area within the mountainous watershed, and the suitable afforestation areas for mountainous tree species are output, including:

[0039] The model is selected as either a statistical model or an artificial intelligence model; wherein the statistical model includes any one of the following: generalized linear model, generalized additive model, multivariate adaptive spline smoothing function model, and flexible discriminant analysis model; the artificial intelligence model includes any one of the following: decision tree and ensemble learning-based model, artificial neural network-based model, support vector machine model, Bayesian model, and maximum entropy model.

[0040] The model is evaluated using cross-validation, random partitioning of sample data, or leave-one-out method;

[0041] The predictive performance is characterized by one or more of the following: area under the acceptor curve, true skill statistics, total accuracy, sensitivity, specificity, omission error, redundancy error, or kappa value. After the model is evaluated, the spatial attribute raster data is used as the predictor variable to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species.

[0042] A system for establishing afforestation areas for mountainous tree species containing bamboo, comprising:

[0043] A mountainous area tree species distribution point data acquisition unit is used to acquire mountainous area tree species distribution point data; the mountainous area tree species distribution point data carries geographic coordinates.

[0044] The watershed boundary data acquisition and delineation unit is used to acquire watershed boundary data. When the watershed boundary data cannot be acquired, it acquires digital elevation data covering the area where the distribution point data of the mountain tree species is located, and delineates the watershed boundary based on the digital elevation data to form the watershed boundary data.

[0045] The mountainous watershed range determination unit is used to select one or more watersheds from the watershed boundary data based on the mountainous tree species distribution point data and combine them to obtain the mountainous watershed range.

[0046] The mountainous watershed topographic map data acquisition unit is used to acquire mountainous watershed topographic map data based on the mountainous watershed area; the mountainous watershed topographic map data is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal;

[0047] The spatial attribute raster data extraction unit is used to extract spatial attribute raster data based on the digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index;

[0048] The training sample set construction unit is used to extract corresponding spatial attribute values ​​from the spatial attribute raster data according to the geographic coordinates of the distribution point data of the mountain tree species, and form a training sample set with the distribution point data of the mountain tree species as the response variable and the spatial attribute values ​​as the prediction variable.

[0049] The afforestation area prediction and output unit is used to build a model based on the training sample set and evaluate the predictive performance of the model. After the model passes the evaluation, it uses the spatial attribute raster data to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species. The model can be either a statistical model or an artificial intelligence model.

[0050] The present invention discloses the following technical effects:

[0051] This invention establishes a model using tree species distribution point data in mountainous areas as the response variable and spatial attribute raster data directly extracted from digital elevation data as the predictor variable. This enables quantitative prediction of suitable afforestation areas for tree species in mountainous regions based solely on topographic information. Because digital elevation data is characterized by continuous coverage, high spatial resolution, and ease of acquisition in mountainous areas, this invention avoids reliance on high-density meteorological and soil sampling data. It effectively reduces prediction bias caused by sparse data points, interpolation uncertainties, and the failure to characterize micro-topographic differences, thereby improving the stability and reliability of afforestation area determination results under complex terrain conditions.

[0052] This invention selects and combines one or more watersheds at the watershed scale to form a mountainous watershed area. Within this area, it uniformly extracts spatial attributes such as altitude, slope, aspect conversion, potential solar radiation, and hydrological data. A training sample set is constructed based on tree species distribution points, and the model's predictive performance is evaluated before spatial prediction is implemented. This provides a clear data constraint and verification mechanism for the establishment of afforestation areas. This technical solution not only maintains spatial consistency and ecological unit integrity in the prediction results at the watershed scale but also allows for the selection of afforestation area distribution results suitable for the target tree species through model evaluation. This enhances the interpretability and applicability of afforestation area delineation in mountainous areas, providing more targeted technical support for scientific afforestation and ecological restoration in mountainous regions. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart of the method provided in an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the implementation route provided in the embodiments of the present invention;

[0056] Figure 3 This is a schematic diagram illustrating the implementation steps of an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram showing the distribution of tree species and the spatial distribution of rivers in the mountainous area within the study area, provided for an embodiment of the present invention.

[0058] Figure 5 This is a contour map of the digital elevation data of the study area provided in an embodiment of the present invention;

[0059] Figure 6 Spatial distribution map of topographic humidity index of the study area provided for embodiments of the present invention;

[0060] Figure 7 Spatial distribution map of heat load index in the study area provided for embodiments of the present invention;

[0061] Figure 8 Spatial distribution map of topographic location index of the study area provided in this embodiment of the invention;

[0062] Figure 9 This is a schematic diagram of the MAXENT maximum entropy model's running interface provided in an embodiment of the present invention;

[0063] Figure 10 This is a schematic diagram illustrating the prediction results of suitable afforestation areas for mountainous tree species in the study area provided in an embodiment of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] The purpose of this invention is to provide a method and system for establishing afforestation areas in mountainous areas containing bamboo species. By integrating tree species distribution points and topographic spatial attributes at the watershed scale and performing spatial prediction after model evaluation, the reliability and applicability of the results of establishing afforestation areas in mountainous areas are improved.

[0066] In this invention, the term "tree species" is used to refer to afforestation objects, and its scope is not limited. It includes both conventional forestry tree species and bamboo species corresponding to bamboo plants. This invention does not impose substantial restrictions on the specific types of afforestation objects. Any afforestation object that can be subjected to suitability analysis using the methods of this invention falls within the protection scope of this invention.

[0067] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] Example 1:

[0069] Figure 1 A flowchart of the method provided in the embodiments of the present invention, such as Figure 1 As shown, this embodiment 1 provides a method for establishing afforestation areas for mountainous tree species containing bamboo, including:

[0070] Step 100: Obtain tree species distribution point data in mountainous areas; the tree species distribution point data in mountainous areas carries geographic coordinates;

[0071] Step 200: Obtain watershed boundary data. If watershed boundary data cannot be obtained, obtain digital elevation data of the area where the tree species distribution data in the mountainous area is located, and delineate the watershed boundary based on the digital elevation data to form watershed boundary data.

[0072] Step 300: Based on the distribution data of tree species in mountainous areas, select one or more watersheds in the watershed boundary data and combine them to obtain the watershed range in mountainous areas;

[0073] Step 400: Obtain topographic map data of the mountainous watershed based on the scope of the mountainous watershed; the topographic map data of the mountainous watershed is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal;

[0074] Step 500: Extract spatial attribute raster data based on digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index;

[0075] Step 600: Based on the geographic coordinates of the distribution point data of tree species in mountainous areas, extract the corresponding spatial attribute values ​​from the spatial attribute raster data to form a training sample set with the distribution point data of tree species in mountainous areas as the response variable and the spatial attribute values ​​as the prediction variable.

[0076] Step 700: Build a model based on the training sample set and evaluate the predictive performance of the model. After the model passes the evaluation, use spatial attribute raster data to predict the watershed area in the mountainous area and output the suitable afforestation area for tree species in the mountainous area. The model can be either a statistical model or an artificial intelligence model.

[0077] like Figure 2 and Figure 3 As shown in the figure, the method for establishing a mountainous afforestation area containing bamboo species according to this application mainly includes the following steps:

[0078] Step S1: Obtaining tree species distribution data in mountainous areas

[0079] The distribution data of tree species in mountainous areas must have clear geographical coordinates and a location accuracy of at least 30 meters. Data sources include: (1) field surveys, (2) forest resource inventory FIA data or forest map data from the National Forestry and Grassland Administration, (3) vegetation maps or remote sensing maps, (4) published literature (monographs, papers) data, such as Web of Science, CNKI, etc.; (5) online databases (such as the Global Biodiversity Platform GBIF, China Digital Herbarium CVH, etc.).

[0080] Step S2, Delineation of watershed area in mountainous areas

[0081] First, obtain the boundary data of each level of watershed. Then, using the distribution range of mountain tree species in step S1, select the required watershed, ensuring that each distribution point falls within one or more selected watershed combinations. The boundary range and area determined by these one or more watershed combinations are considered the final watershed range. To ensure the reliability of subsequent results, try to keep the total watershed area within 1000 square kilometers.

[0082] Watershed boundary data is obtained through free online sharing of watershed boundary data, or by delineating the boundaries of watersheds at all levels through digital elevation data.

[0083] Free sources of shared watershed boundary data include, but are not limited to: (1) global watershed boundary data HydroBASINS; (2) some watershed boundary data shared by the National Basic Science Data Center; and (3) digitization of paper maps showing the distribution of watersheds in the region.

[0084] Digital elevation data is used to delineate the boundaries of watersheds at all levels: the boundaries of watersheds are extracted using the "hydrological analysis" module in software such as geographic information systems.

[0085] To ensure the accuracy of watershed boundary extraction as much as possible, preferred methods are to use digital elevation data that has been corrected for vegetation and building heights. These data sources include, but are not limited to, SGS-UB DTM from the University of Bristol, UK, MERIT DTM from the Japan Agency for Marine-Geoscience and Technology, and FABDEM+.

[0086] Step S3: Acquisition of topographic map data for mountainous watersheds

[0087] Mountainous watershed topographic map data is obtained through digitizing paper topographic maps, applying for topographic map data from national and local natural resource management departments, or downloading digital elevation data that has been corrected for vegetation and building heights by scientific researchers free of charge.

[0088] Paper topographic maps can be obtained by applying to national and local natural resources management departments at all levels.

[0089] The digitization of paper topographic maps can be achieved through the "georegistration" module of professional software tools such as Geographic Information Systems.

[0090] The corrected digital elevation data comes from free, scholarly corrections shared online; see step S2 for the source.

[0091] Digital topographic maps can be converted into digital elevation data using the "topography to raster" function of software such as Geographic Information Systems.

[0092] Step S4: Acquisition of topographic and geomorphological data for mountainous watersheds

[0093] This is achieved by extracting various spatial attribute raster data reflecting the topography and biological processes of mountainous areas from the raster digital elevation data described in step S3. The spatial attributes include, but are not limited to, the following types: altitude, longitude, latitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, plane curvature, profile curvature, total curvature, topographic exposure index, topographic protection index, topographic location index, topographic humidity index, etc.

[0094] The elevation space raster data comes directly from the digital elevation data obtained in step S4.

[0095] Methods for acquiring longitude and latitude spatial raster data: Use the "raster to point" function of software such as Geographic Information System to convert digital elevation data into point vector data, then extract the longitude and latitude coordinates corresponding to each point, and use the "point to raster" function to convert the point vector data into spatial raster longitude and latitude data, while ensuring that the pixel size, range and coordinate system are consistent with the original elevation data.

[0096] Slope spatial raster data is obtained by extracting slope raster data from digital elevation data using the "slope" function of software such as Geographic Information Systems (GIS). Slope represents the degree of inclination of a local surface slope.

[0097] Cosine-to-sine and sine-to-slope aspect spatial raster data acquisition: Slope aspect raster data is extracted from digital elevation data using the "slope aspect" function of software such as Geographic Information Systems (GIS), and then the cosine and sine functions of the slope aspect are calculated respectively. The calculation formulas are as follows:

[0098]

[0099] In the formula, The cosine-to-aspect ratio data is converted, and the converted values ​​range from [-1, 1]. The south slope (180°) corresponds to a cosine value of -1, and the north slope (0°) corresponds to a cosine value of +1. The sine-converted slope aspect data has a value range of [-1, 1], with +1 corresponding to due east (90°) and -1 corresponding to due west (270°); Aspect is the slope aspect angle (0°~360°), with 0 degrees for due north and calculated clockwise, with a value range of 0° to 360°, representing the direction of the maximum change in elevation value at that point.

[0100] The calculation formula for terrain radiation direction raster data is as follows:

[0101]

[0102] In the formula, TRA represents the topographic radiation direction, with a value range of [0, 1]; where "0" indicates a slope aspect facing northeast (typically the coldest and wettest direction in the landscape), and "1" indicates a warmer and drier southwestern slope. Slope is the slope raster data; Aspect is the slope aspect raster data. If no slope aspect is specified, the value of TRA is set to 0.5.

[0103] Potential solar radiation is obtained from digital elevation data through the "solar radiation point" function of software such as geographic information systems.

[0104] The shadow index raster data is extracted from digital elevation data using the "mountain shadow" function in software such as Geographic Information Systems.

[0105] Raster data of planar curvature, profile curvature, and total curvature are extracted from digital elevation data using the "curvature" function in software such as Geographic Information Systems (GIS). Planar curvature and profile curvature describe the bending and variation of the Earth's surface along the horizontal and vertical directions, respectively. Total curvature displays the shape or curvature of the slope.

[0106] Heat load index raster data indicates that southwest-facing slopes should be warmer than southeast-facing slopes, even if they receive the same amount of solar radiation. The calculation formula is as follows:

[0107]

[0108] In the formula, HLI is the heat load index, with a value range of [0, 1], where 0 represents the coldest slope (northeast) and 1 represents the warmest slope (southwest); Aspect is the slope angle (0°~360°).

[0109] Terrain shadow index raster data is extracted from digital elevation data using the "mountain shadow" function in software such as Geographic Information Systems.

[0110] The terrain exposure index raster data was re-aspected to the north-south axis and weighted according to the steepness of the slope. The calculation formula is as follows:

[0111]

[0112] In the formula, SEI is the terrain exposure index; Slope is the slope raster data; and Aspect is the slope aspect raster data.

[0113] The Terrain Conservation Index (TCI) raster data indicates the degree to which a point is protected by the surrounding terrain within a specified neighborhood. Positive values ​​emphasize the convexity of the terrain, while negative values ​​emphasize its concaveness. A negative value indicates no protection, and a positive value indicates protection. It is extracted from digital elevation data using the "Terrain Conservation Index" function in software such as the SAGA Geographic Information System.

[0114] Topographic location index raster data is extracted from digital elevation data using the "Focus Statistics" function in software such as Geographic Information Systems (GIS). The topographic location index compares the elevation of each cell in the digital elevation data with the average elevation of its surrounding specific neighborhood. A positive value indicates that the center point is located above the surrounding average, while a negative value indicates that the center point is located below the surrounding average.

[0115] Accumulated flow raster data is extracted from digital elevation data using the "flow accumulation" function of software such as geographic information systems.

[0116] The runoff intensity index, a raster data measure, measures the erosive power of surface water flow as a function of local slope and upstream catchment area. The calculation formula is as follows:

[0117]

[0118] In the formula, SPI is the sluice gate intensity index, and the larger the SPI value, the greater the erosion risk; AC is the cumulative flow raster data; Area is the raster cell area (m²). 2 ); Slope is the slope raster data.

[0119] The topographic moisture index raster data characterizes the impact of local topography on soil moisture. It is calculated as the natural logarithm of the ratio of the upslope-contributed watershed area to the grid cell slope. The calculation formula is as follows:

[0120]

[0121]

[0122] In the formula, TWI is the topographic humidity index, with low values ​​typically indicating arid and nutrient-poor locations, and high values ​​indicating nutrient-poor and humid locations; FAC is the catchment area raster data; AC is the cumulative flow raster data; and Area is the cell area (m²). 2 ); Slope, slope raster data.

[0123] The topographic convergence index raster data characterizes the degree to which surrounding raster cells converge towards the central raster cell. The convergence index is an effective tool for analyzing topography, such as ridges or waterway systems, and for identifying valleys. Results are expressed as a percentage; negative values ​​correspond to converging flow conditions, and positive values ​​correspond to diverging flow conditions. -100 represents the apex of a cone, 100 represents a depression, and 0 represents a uniform slope. It is extracted from digital elevation data using the "Topographic Convergence Index" function in software such as the GIS SAGA.

[0124] Step S5: Establishment of suitable afforestation areas for mountainous tree species

[0125] Using the mountain tree species distribution data described in step S1 as the response variable (1, tree species exists; 0, does not exist), the topographic and geomorphic attribute data corresponding to the geographical location of the distribution point described in step S4 are extracted as the prediction variable to establish a machine learning / artificial intelligence or statistical model; then the prediction performance of the model is evaluated; finally, using the watershed topographic spatial raster data obtained in step S4 as the prediction variable, the model is used to predict the areas suitable for afforestation within the mountain watershed established in step S2.

[0126] Environmental data acquisition corresponding to distribution points: Using the "Extract Values ​​to Points" function module in software such as Geographic Information System or R language, extract the topographic and geomorphological attribute data values ​​of the corresponding locations based on the geographic coordinates of the distribution points.

[0127] Traditional statistical methods include, but are not limited to, the following types: generalized linear models, generalized additive models, multivariate adaptive regression splines, and flexible discriminant analysis.

[0128] Machine learning / artificial intelligence algorithms include, but are not limited to, the following types: various algorithms based on decision trees and ensemble learning, various algorithms based on artificial neural networks, support vector machines, Bayesian algorithms, maximum entropy models, etc.

[0129] Model evaluation methods include, but are not limited to, the following types: cross-validation, random splitting of sample data, leave-one-out, etc.

[0130] Model prediction performance evaluation metrics include, but are not limited to, the following: Area Under the Acceptance Machine Curve (AUC), True Skill Statistic (TSS), Total Accuracy (OA), Sensitivity (R), Specificity (Spe), Omission Error (OE), Redundancy Error (CE), and kappa value. The error matrix used for model prediction performance evaluation is shown in Table 1.

[0131] Table 1. Error matrix used to evaluate model prediction accuracy

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139]

[0140] In the formula, OA is the total precision; Sen is the sensitivity; Spe is the specificity; OE is the omission error; CE is the redundancy error; TSS is the true skill statistical method; and n is the total number of samples.

[0141] In this embodiment, in the step of building a model based on the training sample set, the model is selected as the maximum entropy model MAXENT, as follows:

[0142] First, feature mapping is performed on the spatial attribute values ​​corresponding to the distribution points of each mountain tree species in the training sample set to form a sample feature representation usable by MAXENT; specifically, for the first... Feature vectors were constructed from the distribution data of the tree species in the mountainous areas. Each component corresponds to at least one spatial attribute value extracted from the spatial attribute raster data, including altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index, and the samples with a response variable of 1 in the training sample set are used as the existing sample set.

[0143]

[0144] in, The number of existing samples; For feature dimensions; For the first The feature vector of an existing sample; for The Dimensional component, corresponding to the first Item space attribute value; For the first One characteristic function; For the existence of samples in the th... The empirical expectation under each feature function. After completing the feature mapping, within the mountainous watershed area, using grid cells as the prediction object, let... This represents the set of all raster cells within the mountainous watershed area. For any raster cell... Extract its corresponding spatial attribute feature vector MAXENT obtains the probability distribution form of grid cells by maximizing the probability distribution entropy under the condition of satisfying the "feature expectation consistency constraint", and outputs the suitability prediction result accordingly:

[0145]

[0146] in, For grid cells The corresponding maximum entropy probability; A collection of raster cells within a mountainous watershed; For grid cells eigenvectors; For the first The coefficients of the characteristic functions; This is the normalization factor.

[0147] Furthermore, the parameters of the maximum entropy model MAXENT are determined by satisfying the constraint that the expected value of existing samples is consistent with the expected value of the model. Thus, without introducing additional environmental surface data, the constraints and objectives of characterizing the impact of micro-topographical differences in mountainous areas on tree species suitability based solely on the spatial attribute raster data can be expressed as follows:

[0148]

[0149] in, For probability distribution Entropy; It is the natural logarithm function; For the existence of samples in the th... Empirical expectation under each characteristic function.

[0150] After the model passes the prediction performance evaluation, the spatial attribute raster data is input into the maximum entropy model MAXENT. A suitability score is output for each raster cell within the mountainous watershed area, and this suitability score is used as the spatial prediction result for suitable afforestation areas for the mountainous tree species. To facilitate the formation of suitability raster data that can be directly used for mapping, this embodiment uses the logarithmic linear term corresponding to the probability distribution as a continuous suitability score and performs a monotonic transformation.

[0151]

[0152] in, For grid cells Linear scoring; For grid cells The suitability score, with a value range of ; It is an exponential function; the other symbols are the same as those in the previous formula.

[0153] This embodiment preferably uses the MAXENT maximum entropy model as the model. By mapping the spatial attribute values ​​to feature functions and obtaining the probability distribution of each grid cell within the mountainous watershed under the constraint of feature expectation consistency, a robust estimation of suitability distribution can be achieved using only the mountainous tree species distribution point data and the spatial attribute grid data. Compared with methods that mainly fit fixed thresholds or simple linear relationships, the maximum entropy model selects the probability distribution form with the least information under the premise of consistent constraints. This can reduce the estimation bias caused by limited sample size and strong terrain heterogeneity, making the suitability prediction results more stable and insensitive to slight perturbations of training samples.

[0154] Furthermore, this embodiment outputs a continuous suitability score based on MAXENT's log-linear scoring, which allows suitable afforestation areas to be expressed not only in a zonal form but also in a continuous gradient manner to express the strength and weakness of habitat suitability, thereby improving the interpretability and usability of the results. At the same time, the characteristic function and its corresponding coefficient can reflect the contribution direction and relative influence of each spatial attribute on suitability changes, which facilitates the interpretation and analysis of the prediction results and supports the decision-making on afforestation area establishment.

[0155] Example 1: Taking the mountainous tree species, thick-branched spruce, as an example, the experimental process and results of this example are as follows:

[0156] In step S1, data on 39 distribution points of thick-branched spruce were obtained using the field survey method. Figure 4 );

[0157] In step S2, by downloading FABDEM+ data, the "Hydrological Analysis" module of the geographic information system is used to delineate the watershed range at each level. Combined with the distribution point data of spruce in step S1, the final analysis watershed range is delineated.

[0158] In step S3, contour line data is obtained by digitizing the paper topographic map. This data is then cropped according to the watershed area defined in step S2, and the cropped result is used as the final watershed topographic map data.

[0159] Step S4: Using the watershed topographic map data obtained in step S3 as analysis data, obtain raster data of topographic and geomorphic process attributes of the mountainous watershed, including: altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, plane curvature, profile curvature, total curvature, topographic exposure index, topographic protection index, topographic location index, and topographic humidity index.

[0160] Step S5: Establish a prediction model for suitable afforestation areas of *Picea gracilis* based on the MAXENT maximum entropy model to obtain the location of suitable afforestation sites for the target afforestation area. The prediction variables input to the model include all spatial raster data of various topographic attributes obtained in step S4, and the response variables input to the model are binary classification data of the presence (1) or absence (0) of *Picea gracilis*. Default parameter settings are used during model training. The model is evaluated using a random data partitioning method, i.e., 70% of the data is used as model training data, and the remaining 30% is used as model evaluation data. The area AUC and omission error under the receiver operating characteristic curve are used to evaluate the model. The MAXENT model prediction results are converted into binary prediction values ​​(1, suitable afforestation area; 0, unsuitable afforestation area) using the threshold corresponding to the maximum sum of sensitivity and specificity.

[0161] Figure 4 This is a schematic diagram of the distribution points of tree species and the spatial distribution of rivers in the mountainous area within the study area provided in this embodiment of the invention. Figure 4 As shown, Figure 4 The spatial extent of the study area is represented by a planar boundary, with black dots indicating the acquired distribution data of tree species in the mountainous area, each carrying its corresponding geographic coordinates. Blue lines represent the spatial distribution of rivers within the study area. This figure visually reflects the spatial relationships of tree species distribution points within a watershed scale, providing a spatial reference for subsequently selecting one or more watersheds from the watershed boundary data based on tree species distribution points and combining them to obtain the scope of the mountainous watershed.

[0162] Figure 5 This is a contour map of the digital elevation data of the study area provided in an embodiment of the present invention, such as... Figure 5 As shown, Figure 5The topographic relief features generated from digital elevation data are expressed in the form of contour lines. These contour lines are continuously distributed within the study area, reflecting the overall elevation structure and topographic complexity of the mountainous terrain. This digital elevation data, serving as topographic map data for a mountainous watershed, provides fundamental topographic information for subsequent extraction of raster data related to slope, aspect, runoff, and humidity.

[0163] Figure 6 The spatial distribution map of the topographic humidity index of the study area provided in this embodiment of the invention is as follows: Figure 6 As shown, Figure 6 The spatial distribution of the topographic humidity index, calculated based on digital elevation data, within the study area is displayed in raster form, with different colors corresponding to different humidity levels. This topographic humidity index reflects the influence of topographic conditions on water collection and retention capacity and is an important component of spatial attribute raster data, used to characterize the differences in water environment at different locations in mountainous areas.

[0164] Figure 7 The spatial distribution map of the heat load index of the study area provided in the embodiments of the present invention is as follows: Figure 7 As shown, Figure 7 The spatial distribution of the heat load index, calculated by comprehensively considering factors such as slope, aspect, and potential solar radiation, within the study area is represented in grid form, with different colors indicating the intensity of the heat load. This heat load index is used to characterize the modulating effect of topographic conditions on the distribution of solar radiation and heat, reflecting the differences in the thermal environment at different topographic locations.

[0165] Figure 8 The spatial distribution map of the topographic location index of the study area provided in this embodiment of the invention is as follows: Figure 8 As shown, Figure 8 The spatial distribution of topographic location indices within the study area is displayed in raster form. These indices reflect the relative elevation of each raster cell with respect to the surrounding terrain, and are used to distinguish different terrain types such as ridges, slopes, and valleys. This index helps characterize micro-topographic differences and is an integral part of the spatial attribute raster data.

[0166] Figure 9 This is a schematic diagram of the MAXENT maximum entropy model's running interface provided in an embodiment of the present invention, as shown below. Figure 9 As shown, Figure 9 The document showcases the interface for the MAXENT maximum entropy model, including sample data input areas, environmental variable input areas, and model output parameter settings. This interface allows users to input data on the distribution points of tree species in mountainous areas as samples and spatial attribute raster data as environmental variables, performing model training and prediction to provide model support for subsequently outputting suitable afforestation areas for tree species in mountainous regions.

[0167] Figure 10This is a schematic diagram illustrating the prediction results of suitable afforestation areas for mountainous tree species in the study area provided in this embodiment of the invention. Figure 10 As shown, Figure 10 The figure presents the spatial distribution of suitable afforestation areas for mountainous tree species predicted by the model in a planar distribution format. The dark areas represent the predicted suitable afforestation areas, while the remaining areas represent unsuitable or less suitable areas. The distribution points of tree species and river distributions are also overlaid to compare and analyze the spatial consistency between the predicted results and existing distribution information. This figure is a visual representation of the final output of the method of this invention.

[0168] Example 2:

[0169] A method for establishing afforestation areas for mountainous tree species containing bamboo is identical to that in Example 1, except for step S5, which involves "predicting suitable afforestation sites for coarse-branched spruce through random forest algorithm modeling".

[0170] Example 3:

[0171] A method for establishing afforestation areas in mountainous areas containing bamboo species is the same as in Example 1, except that in step S5, "the number of predictive variables used for modeling is pre-selected by variable importance".

[0172] Corresponding to the above method, the present invention also provides a system for establishing afforestation areas for mountainous tree species containing bamboo, comprising:

[0173] A mountainous area tree species distribution point data acquisition unit is used to acquire mountainous area tree species distribution point data; the mountainous area tree species distribution point data carries geographic coordinates.

[0174] The watershed boundary data acquisition and delineation unit is used to acquire watershed boundary data. When the watershed boundary data cannot be acquired, it acquires digital elevation data covering the area where the distribution point data of the mountain tree species is located, and delineates the watershed boundary based on the digital elevation data to form the watershed boundary data.

[0175] The mountainous watershed range determination unit is used to select one or more watersheds from the watershed boundary data based on the mountainous tree species distribution point data and combine them to obtain the mountainous watershed range.

[0176] The mountainous watershed topographic map data acquisition unit is used to acquire mountainous watershed topographic map data based on the mountainous watershed area; the mountainous watershed topographic map data is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal;

[0177] The spatial attribute raster data extraction unit is used to extract spatial attribute raster data based on the digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index;

[0178] The training sample set construction unit is used to extract corresponding spatial attribute values ​​from the spatial attribute raster data according to the geographic coordinates of the distribution point data of the mountain tree species, and form a training sample set with the distribution point data of the mountain tree species as the response variable and the spatial attribute values ​​as the prediction variable.

[0179] The afforestation area prediction and output unit is used to build a model based on the training sample set and evaluate the predictive performance of the model. After the model passes the evaluation, it uses the spatial attribute raster data to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species. The model can be either a statistical model or an artificial intelligence model.

[0180] The beneficial effects of this invention are as follows:

[0181] (1) This invention establishes a model and outputs suitable afforestation areas by using mountain tree species distribution point data as response variables and spatial attribute raster data extracted from digital elevation data as spatial prediction variables. This makes the process of establishing afforestation areas no longer dependent on the areal interpolation results of climate data and soil data, thereby avoiding the problem of amplified interpolation errors caused by sparse monitoring points in mountainous areas and drastic changes in the environment over short distances, and improving the stability and reliability of the prediction of suitable afforestation areas.

[0182] (2) When it is not possible to directly obtain watershed boundary data, the present invention can obtain digital elevation data of the area where the tree species distribution points are located and delineate the watershed boundary to form watershed boundary data. Then, based on the tree species distribution points, the watershed is selected and combined in the watershed boundary data to obtain the watershed range of the mountainous area. This allows subsequent data extraction and prediction to be carried out within a clear spatial boundary, reducing the result deviation caused by the arbitrariness of the research range selection and improving the spatial consistency and reproducibility of the afforestation area establishment.

[0183] (3) Based on the watershed range of the mountainous area, the present invention obtains digital elevation data as topographic map data of the watershed, and extracts spatial attribute raster data such as altitude, slope, aspect conversion, potential solar radiation, shadow index and confluence and humidity related attributes from the digital elevation data, so that the prediction variables can directly reflect the differences in hydrothermal conditions and micro-topographic microhabitat characteristics under the control of topography, thereby improving the ability to characterize the differences in habitat suitability of tree species under complex topographic conditions in mountainous areas and spatial differentiation.

[0184] (4) According to the geographical coordinates of the tree species distribution points, the present invention extracts the corresponding spatial attribute values ​​from the spatial attribute raster data, constructs a training sample set that matches the response variable and the prediction variable, and forms a clear data transfer relationship of "point sample, raster attribute, model training". This avoids sample mismatch caused by inconsistent spatial benchmarks of multi-source data and improves the effectiveness of model training data and the reliability of prediction results.

[0185] (5) After the model is established, the prediction performance is evaluated. After the model passes the evaluation, spatial attribute raster data is used to perform spatial prediction within the mountainous watershed and output suitable afforestation areas. This makes the output results have evaluation basis and verifiability, which facilitates the selection of applicable models under different tree species and different mountainous watershed conditions and obtains stable results of suitable afforestation area division, thereby providing more targeted technical support for scientific afforestation, ecological restoration and resource management in mountainous areas.

[0186] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0187] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for establishing afforestation areas in mountainous regions containing bamboo species, characterized in that, include: Acquire data on the distribution points of tree species in mountainous areas; the data on the distribution points of tree species in mountainous areas carries geographical coordinates. Obtain watershed boundary data; if the watershed boundary data cannot be obtained, obtain digital elevation data covering the area where the distribution point data of the mountain tree species is located, and delineate the watershed boundary based on the digital elevation data to form the watershed boundary data. Based on the distribution data of tree species in the mountainous area, one or more watersheds are selected from the watershed boundary data and combined to obtain the watershed range of the mountainous area; Topographic map data of the mountainous watershed is obtained based on the watershed area of ​​the mountainous area; the topographic map data of the mountainous watershed is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal; Spatial attribute raster data is extracted based on the digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index; Based on the geographic coordinates of the mountain tree species distribution point data, corresponding spatial attribute values ​​are extracted from the spatial attribute raster data to form a training sample set with the mountain tree species distribution point data as the response variable and the spatial attribute values ​​as the prediction variable. A model is established based on the training sample set, and the predictive performance of the model is evaluated. After the model passes the evaluation, the spatial attribute raster data is used to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species. The model can be either a statistical model or an artificial intelligence model.

2. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Obtain data on the distribution points of tree species in mountainous areas, including: Data on the distribution points of tree species in mountainous areas with geographic coordinates are obtained, and the location accuracy of the geographic coordinates is at least 30 meters; the data sources for the distribution points of tree species in mountainous areas include at least one or more of the following: field survey data, forest resource inventory data or forest resource data released by forest resource management departments, vegetation maps or remote sensing map data, published literature data, and online database data.

3. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Obtain watershed boundary data, including: Obtain free online shared watershed boundary data, or digitize a paper map of the watershed distribution in the region to obtain the watershed boundary data; wherein, the free online shared watershed boundary data includes at least the watershed boundary data shared by the global watershed boundary data HydroBASINS or the National Basic Science Data Center.

4. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... When the watershed boundary data cannot be obtained, digital elevation data covering the area where the distribution data of the tree species in the mountainous area is located is obtained, and the watershed boundary is delineated based on the digital elevation data to form the watershed boundary data, including: The watershed extent is extracted from the digital elevation data using the hydrological analysis module of the geographic information system software to form the watershed boundary data; wherein, the digital elevation data is the digital elevation data after vegetation and building height correction and removal.

5. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Based on the distribution data of tree species in the mountainous area, one or more watersheds are selected from the watershed boundary data and combined to obtain the range of the mountainous watershed, including: One or more watersheds are selected based on the geographic coordinates of the mountain tree species distribution point data, such that each of the mountain tree species distribution point data falls within the combined boundary of the one or more watersheds. The combined boundary range of the one or more watersheds is defined as the range of the mountainous watershed, and the total area of ​​the watershed corresponding to the range of the mountainous watershed is controlled within 1,000 square kilometers.

6. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Based on the aforementioned mountainous watershed area, topographic map data of the mountainous watershed is obtained, including: Obtain digital elevation data as the topographic map data of the mountainous watershed, wherein the digital elevation data is obtained in any of the following ways: Perform georegistration on paper topographic maps to achieve digitization, and obtain the digital elevation data through the terrain-to-raster conversion function; or Apply for topographic map data from national and local natural resources management departments at all levels, and obtain the digital elevation data from the topographic map data; or Download the digital elevation data after vegetation and building height correction and removal, and use the corrected and removed digital elevation data as the topographic map data of the mountainous watershed.

7. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Extracting spatial attribute raster data based on the digital elevation data includes: The slope is extracted from the digital elevation data using the slope function; The slope aspect raster data is extracted from the digital elevation data using the slope aspect function, and cosine and sine function operations are performed on the slope aspect raster data respectively to obtain the cosine-converted slope aspect and the sine-converted slope aspect; The potential solar radiation is extracted from the digital elevation data using the solar radiation point function; The shadow index is extracted from the digital elevation data using the mountain shadow function; Accumulated flow raster data is extracted from the digital elevation data using the flow accumulation function to generate the runoff area, the runoff intensity index, and the topographic wetness index.

8. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... Based on the geographic coordinates of the mountain tree species distribution point data, corresponding spatial attribute values ​​are extracted from the spatial attribute raster data to form a training sample set with the mountain tree species distribution point data as the response variable and the spatial attribute values ​​as the prediction variable, including: The distribution data of tree species in the mountainous area is used as the response variable, and the response variable is set as a binary variable, where 1 indicates that the tree species exists and 0 indicates that the tree species does not exist. Using the "Extract Values ​​to Points" function module in Geographic Information System software or R language software, the corresponding spatial attribute values ​​are extracted from the spatial attribute raster data according to the geographic coordinates of the distribution points of tree species in the mountainous area, and used as the prediction variables; The response variable is paired with the predictor variable to form the training sample set.

9. The method for establishing afforestation areas for mountainous tree species containing bamboo, as described in claim 1, is characterized in that... A model is built based on the training sample set, and the predictive performance of the model is evaluated. After the model passes the evaluation, the spatial attribute raster data is used to predict the suitable afforestation areas for mountainous watersheds, and the results are output, including: The model is selected as either a statistical model or an artificial intelligence model; wherein the statistical model includes any one of the following: generalized linear model, generalized additive model, multivariate adaptive spline smoothing function model, and flexible discriminant analysis model; the artificial intelligence model includes any one of the following: decision tree and ensemble learning-based model, artificial neural network-based model, support vector machine model, Bayesian model, and maximum entropy model. The model is evaluated using cross-validation, random partitioning of sample data, or leave-one-out method; The predictive performance is characterized by one or more of the following: area under the acceptor curve, true skill statistics, total accuracy, sensitivity, specificity, omission error, redundancy error, or kappa value. After the model is evaluated, the spatial attribute raster data is used as the predictor variable to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species.

10. A system for establishing afforestation areas for mountainous tree species containing bamboo, characterized in that, include: A mountainous area tree species distribution point data acquisition unit is used to acquire mountainous area tree species distribution point data; the mountainous area tree species distribution point data carries geographic coordinates. The watershed boundary data acquisition and delineation unit is used to acquire watershed boundary data. When the watershed boundary data cannot be acquired, it acquires digital elevation data covering the area where the distribution point data of the mountain tree species is located, and delineates the watershed boundary based on the digital elevation data to form the watershed boundary data. The mountainous watershed range determination unit is used to select one or more watersheds from the watershed boundary data based on the mountainous tree species distribution point data and combine them to obtain the mountainous watershed range. The mountainous watershed topographic map data acquisition unit is used to acquire mountainous watershed topographic map data based on the mountainous watershed area; the mountainous watershed topographic map data is digital elevation data, which is obtained by digitizing paper topographic maps, applying for topographic map data, or downloading digital elevation data after vegetation and building height correction and removal; The spatial attribute raster data extraction unit is used to extract spatial attribute raster data based on the digital elevation data; the spatial attribute raster data includes altitude, potential solar radiation, slope, cosine-converted aspect, sine-converted aspect, runoff area, shadow index, runoff intensity index, and topographic humidity index; The training sample set construction unit is used to extract corresponding spatial attribute values ​​from the spatial attribute raster data according to the geographic coordinates of the distribution point data of the mountain tree species, and form a training sample set with the distribution point data of the mountain tree species as the response variable and the spatial attribute values ​​as the prediction variable. The afforestation area prediction and output unit is used to build a model based on the training sample set and evaluate the predictive performance of the model. After the model passes the evaluation, it uses the spatial attribute raster data to predict the area within the mountainous watershed and output the suitable afforestation area for mountainous tree species. The model can be either a statistical model or an artificial intelligence model.

Images