Urban waterfront green land carbon sequestration benefit key influence factor identification method

Abstract

The invention discloses an urban waterfront green land carbon sequestration benefit key influence factor identification method. Firstly, remote sensing image data, land utilization data and NDVI data of a target area are acquired, and the average carbon sequestration efficiency of each waterfront green land in the target area is calculated by combining LAI actual measurement inversion and tree and shrub actual measurement investigation. Secondly, constructing an influence factor index set; thirdly, calculating target area influence factor index data to obtain an index data set; and finally, on the basis of the influence factor index data set obtained in the above steps, calculating the importance of each index through a random forest regression model to obtain a key factor influencing the carbon sequestration benefit of the urban waterfront green land. According to the method, the key influence factor of the carbon sequestration benefit of the urban waterfront green land can be quickly identified, and a basis can be provided for updating the waterfront green land under a high-carbon sink target.

Description

Technical Field

[0001] The present invention belongs to the technical field of ecological environment quality assessment, and particularly relates to a method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces. Background Art

[0002] Waterfront green spaces are important contributors to the carbon sink in urban areas. Existing studies have shown that the synergy between water and green spaces can be promoted by optimizing their relationship, thereby improving the carbon sequestration benefits of green spaces. Previous studies have mostly focused on the relationship between the spatial form of urban green spaces and their carbon sequestration benefits, lacking a method for efficiently identifying the key influencing factors of carbon sequestration benefits in waterfront green spaces from the perspective of water-green synergy. Therefore, it is necessary to propose an identification method targeting the characteristics of urban waterfront areas to provide support for the renewal of waterfront green spaces under the goal of high carbon sequestration. Summary of the Invention

[0003] To solve the above problems, the present invention discloses a method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces. Aiming at carbon sequestration enhancement, it can quickly identify the key influencing factors affecting the carbon sequestration benefits of waterfront green spaces according to the characteristics of different urban waterfront areas. It is convenient, efficient, accurate in operation, and can provide a basis for the renewal of waterfront green spaces under the goal of high carbon sequestration.

[0004] To achieve the above object, the technical solution of the present invention is as follows:

[0005] A method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces, comprising the following steps:

[0006] S1. Obtain the remote sensing image data, land use data, and NDVI (Normalized Difference Vegetation Index) data of the target area, and calculate the average carbon sequestration efficiency of each waterfront green space in the target area by combining the measured inversion of LAI (Leaf Area Index) and the measured investigation of trees and shrubs.

[0007] S2. Construct an influencing factor index set, including two categories: the morphological index of urban waterfront green spaces and the water-green space relationship index.

[0008] S3. Based on the index set in S2 and the land use data in S1, calculate the data of the morphological index of waterfront green spaces and the water-green space relationship index in the target area to obtain the influencing factor index data.

[0009] S4. Based on the average carbon sequestration efficiency of waterfront green spaces obtained in S1 and the influencing factor index data obtained in S3, construct a regression analysis data set, and calculate the importance of the morphological index of urban waterfront green spaces and the water-green space relationship index through a random forest regression model. The top three or top five in importance are the key factors affecting the carbon sequestration benefits of urban waterfront green spaces.

[0010] Furthermore, in step S1, the specific steps for calculating the average carbon sequestration efficiency of each waterfront green space in the target area are as follows:

[0011] S11. Obtain the remote sensing image data of the target area, and perform image supervised classification in ArcGIS (Arc Geographic Information System) software to obtain the land use data of the target area.

[0012] S12. Based on the remote sensing image data of the target area, calculate the NDVI data of the target area on the ENVI (The Environment for Visualizing Images) platform.

[0013] Among them, the calculation formula is:

[0014]

[0015] Among them, NIR and R respectively correspond to the near-infrared spectrum and red spectrum channels in the remote sensing image data.

[0016] S13. Based on the land use data in step S11, randomly select 30 measured LAI (Leaf Area Index) sample points and 200 tree and shrub survey sample points in the green spaces of the target area through ArcGIS software. Determine the quadrats with the geometric centers of the sample points. The shape of the quadrat is a square or approximately square with a size of 10×10m. Measure the LAI values of trees and shrubs in the quadrats through a canopy analyzer; record the types of trees and shrubs in the tree and shrub survey quadrats to obtain the types of trees and shrubs in different quadrats.

[0017] S14. Taking the LAI values of the quadrats in step S13 as the dependent variable and the NDVI values corresponding to the measured LAI quadrats in S13 as the independent variable, establish a linear regression equation. Combine the NDVI data in step S12 to invert the LAI values of the green spaces in the target area to obtain the LAI data of the green spaces in the target area.

[0018] S15. Combine the types of trees and shrubs in the tree and shrub survey quadrats in step S13, and take the types with higher frequencies as the common trees and shrubs in the green spaces of the target area. Based on the measured carbon sequestration efficiency data of each common tree and shrub, take its average value as the carbon sequestration efficiency of the trees and shrubs in the green spaces of the target area; combine the LAI data of the green spaces in the target area in step S14 to calculate the carbon sequestration efficiency of the green spaces in the target area.

[0019] Among them, the formula for calculating the carbon sequestration efficiency of the green spaces in the target area is:

[0020]

[0021] Among them, C is the daily carbon sequestration efficiency of green space (g·m-2·d-1); y is the LAI value of trees or shrubs; c is the average daily carbon sequestration amount per unit leaf area of the i-th plant type (g·m-2·d-1); t is the effective photosynthetic days of the i-th plant type (d); i is different plant species.

[0022] S16. Based on the remote sensing image data in step S11, in the ArcGIS software, using the road as the boundary, divide the waterfront green space in the target area to obtain the vector range of each waterfront green space in the target area. Combining the green space carbon sequestration efficiency of the target area in S15, in the ArcGIS software, use the "clip" tool to obtain the carbon sequestration efficiency distribution data of each waterfront green space in the target area, and then statistically obtain the average carbon sequestration efficiency of each waterfront green space.

[0023] Furthermore, the influencing factor index set in step S2 includes: 2 waterfront green space morphological indexes, namely the crown coverage rate and the fractal dimension; 5 water-green space relationship indexes, namely the water-green contact degree, the water-green width ratio, the proportion of forest land area within 10m, the forest land within 10 - 20m, and the proportion of forest land area within 20 - 30m.

[0024] Furthermore, in step S3, the specific steps for the influencing factor index data of each waterfront green space are as follows:

[0025] S31. According to the land use data in step S11 and the vector range of each waterfront green space in the target area in S16, in the ArcGIS software, use the "clip" tool to obtain the land use data of each waterfront green space. Subsequently, use the "raster to vector" tool to obtain the vector data of forest land and green space of each waterfront green space.

[0026] S32. According to the vector data of forest land and green space of each waterfront green space in step S31, measure the forest land area and the total green space area in different waterfront green spaces. The formula for calculating the crown coverage rate of the waterfront green space is:

[0027]

[0028] Among them, T i is the crown coverage rate of the i-th waterfront green space, a i is the direct sunlight shadow area of the forest land canopy of the i-th waterfront green space, b i is the total green space area of the i-th waterfront green space.

[0029] S33. According to the vector data of forest land and green space of each waterfront green space in step S31, measure the total green space area and the green space perimeter in each waterfront green space, and calculate the fractal dimension of the green space. The formula is:

[0030]

[0031] Among them, FRAC is the fractal dimension of the green space, and p i is the perimeter of the i-th waterfront green space; a i is the area of the i-th waterfront green space.

[0032] S34. According to the forest land and green space vector data of each waterfront green space in step S31, export them through ArcGIS software, measure the shoreline length of each waterfront green space using CAD software, and combine the perimeter of each waterfront green space measured by ArcGIS software to calculate the degree of water-green contact of the waterfront green space. The formula is:

[0033]

[0034] Among them, R i is the degree of water-green contact of the i-th waterfront green space, L i is the shoreline length of the waterfront green space of the i-th waterfront green space, and P i is the perimeter of the i-th waterfront green space.

[0035] S35. According to the land use data and remote sensing image data in step S11, in ArcGIS software, determine the center lines of the waterfront green space and the river through the binary method; import the waterfront green space, the river channel and their center lines into CAD software, make the perpendicular bisectors of the center lines of the waterfront green space and the corresponding river channel center lines, and use the "path array" tool to perform an array operation with the perpendicular bisector as the object and the center lines of the waterfront green space and the corresponding river channel as the path. Modify the array parameters and adjust them to an appropriate spacing. After cropping, import them into ArcGIS software, and after converting to vector data, obtain the average width of the waterfront green space and the corresponding river channel. The formula for calculating the water-green width ratio is:

[0036]

[0037] Among them, W i is the water-green width ratio of the i-th waterfront green space, W_R i is the average width of the corresponding river section of the i-th waterfront green space, and W_G i is the average width of the i-th waterfront green space.

[0038] S36. According to the forest land and green space vector data of each waterfront green space in step S31 and the land use data in S11, use the "buffer" tool of ArcGIS software to obtain each buffer of the water body, and then use the "clip" tool to obtain the forest land range of the waterfront green space within 10, 10 - 20, and 20 - 30 m from the shoreline, and calculate the proportion of the forest land area within 10, 10 - 20, and 20 - 30 m. The formula is:

[0039]

[0040] Among them, P ijis the ratio of the forest area of the \(i\)-th waterfront green space in different buffer zones to the total forest area of the \(i\)-th green space, \(A\). ij is the forest area of the \(i\)-th waterfront green space in different buffer zones. \(d\). i is the total forest area of the \(i\)-th waterfront green space. \(j\) is the buffer zone at different distances from the shoreline, which are 10m, 10 - 20m, and 20 - 30m respectively.

[0041] Further, in step S4, the specific steps for calculating the importance of each influencing factor of the urban waterfront green space are as follows: First, taking the average carbon sequestration efficiency of the waterfront green space in the target area in step S1 as the dependent variable and the influencing factor index data in the target area in step S3 as the independent variable, a regression analysis dataset is constructed. Second, it is divided into a training set and a test set in a ratio of 7:3, using 5-fold cross-validation, with the maximum number of leaf nodes being 50, the maximum depth of the tree being 10, and the number of decision trees being 100. A random forest model is constructed to analyze the regression analysis dataset to obtain the importance of each factor. Taking \(R\). 2 and the Mean Absolute Error as the necessary conditions for the model to pass the accuracy verification.

[0042] Among them, the calculation formula of \(R\). 2 is as follows:

[0043]

[0044] Among them, \(k\) is the number of folds of cross-validation, that is, the number of times of dividing the training set and the test set. The division ratio for constructing the response model is 0.8, and the value of \(k\) is 5. \(y\). i is the predicted value of the dependent variable for the \(i\)-th calculation, is the actual value of the dependent variable for the \(i\)-th calculation, is the average value of the dependent variable, and \(n\) is the number of samples.

[0045] Second, the calculation formula of the Mean Absolute Error is as follows:

[0046]

[0047] Among them, \(k\) is the number of folds of cross-validation, \(y\). i is the predicted value of the dependent variable, is the actual value of the dependent variable, and \(n\) is the number of samples in the dataset.

[0048] The beneficial effects of the present invention are:

[0049] A method for identifying key influencing factors of carbon sequestration benefits of urban waterfront green spaces provided by the present invention aims at carbon sink enhancement, and can quickly identify key influencing factors affecting the carbon sequestration benefits of waterfront green spaces according to the characteristics of different urban waterfront areas, with the characteristics of convenient operation, high efficiency and accuracy. Description of the Drawings

[0050] Figure 1 is the flowchart of the method for identifying the key influencing factors of the carbon sequestration benefit of urban waterfront green spaces according to the present invention;

[0051] Figure 2 is the specific remote sensing data regional map of the implementation case of the present invention;

[0052] Figure 3 is the distribution map of the carbon sequestration efficiency of the green spaces in the implementation case of the present invention;

[0053] Figure 4 is the measured sample point map of the implementation case of the present invention;

[0054] Figure 5 is the LAI-NDVI regression curve graph of the implementation case of the present invention;

[0055] Figure 6 is the importance ranking map of the influencing factors in the implementation case of the present invention. Detailed Embodiment

[0056] The present invention will be further illustrated below in conjunction with the drawings and the detailed embodiment. It should be understood that the following detailed embodiment is only used to illustrate the present invention and not to limit the scope of the present invention.

[0057] In this embodiment, the Qinhuai New River in Nanjing and the green spaces within 120 m on both sides are used as the target area.

[0058] As Figure 1 shown, a method for identifying the key influencing factors of the carbon sequestration benefit of urban waterfront green spaces according to the present invention specifically includes the following steps:

[0059] S1. Taking the Qinhuai New River in Nanjing and the green spaces within 120 m on both sides as the target area, and the range of the target area is as Figure 2 , obtaining the remote sensing image data, land use data, and NDVI (Normalized Difference Vegetation Index) data of the target area, and calculating the average carbon sequestration efficiency of each waterfront green space in the target area by combining the measured inversion of LAI (Leaf Area Index) and the measured investigation of trees and shrubs. The specific process is as follows:

[0060] S11. Obtaining the remote sensing image data of the target area and performing image supervised classification in the ArcGIS software to obtain the land use data of the target area.

[0061] S12. Calculating the NDVI data of the target area based on the remote sensing image data of the target area in the Envi platform.

[0062] Among them, the calculation formula is:

[0063]

[0064] Among them, NIR and R respectively correspond to the near-infrared spectrum and red spectrum channels in the remote sensing image data.

[0065] S13. Based on the land use data in S11, 30 LAI measured sample points and 200 tree and shrub survey sample points are randomly selected in the green space through ArcGIS software, as Figure 4 . The survey quadrats are determined by the geometric centers of the sample points, and the shape of the survey quadrats is a square or approximately square with a size of 10×10m. In the LAI measured quadrats, the plant canopy analyzer (24MPDSLR Compact OMount) is used to measure the LAI of the trees and shrubs in the quadrats 4 times, and the average value is taken as the LAI value of the LAI measured quadrats. In the tree and shrub survey quadrats, the types of trees and shrubs are recorded to obtain the types of trees and shrubs in different quadrats.

[0066] S14. Taking the quadrat LAI value in S13 as the dependent variable and the NDVI value of the remote sensing image data corresponding to the LAI measured quadrats in S13 as the independent variable, a linear regression equation is established, and the result is as Figure 5 . The regression equation is: y = 5.805x + 0.232, and its standard deviation is 0.607. Combining the NDVI data in S12, the LAI value of the green space in the target area is inverted to obtain the LAI data of the green space in the target area.

[0067] S15. Combining the types of trees and shrubs in the tree and shrub survey quadrats in S13, the types with higher frequencies are taken as the common trees and shrubs in the target area. Based on the measured data of the carbon sequestration efficiency of each common tree and shrub, as shown in Table 1, the average value is taken as the carbon sequestration efficiency of the trees and shrubs in the green space of the target area. Combining the LAI data of the green space in the target area in S14, the carbon sequestration efficiency of the green space in the target area is calculated, and the calculation result is as Figure 3 .

[0068] Table 1 Carbon sequestration capacity of common plants in the target area

[0069]

[0070]

[0071] Among them, the formula for calculating the carbon sequestration efficiency of the green space in the target area is:

[0072]

[0073] Among them, C is the daily carbon sequestration efficiency of the green space (g·m-2·d-1); y is the LAI value of the tree or shrub; c is the average daily carbon sequestration amount per unit leaf area of the i-th plant type (g·m-2·d-1); t is the effective photosynthetic days of the i-th plant type (d); i is different plant species.

[0074] S16. Based on the S11 remote sensing image data, in the ArcGIS software, using roads as boundaries, divide the waterfront green spaces within the target area to obtain the vector ranges of each waterfront green space in the target area. According to the vector ranges of each waterfront green space in the target area in S16 and the carbon sequestration efficiency of the green spaces in the target area in S15, in the ArcGIS software, use the "clip" tool to obtain the carbon sequestration efficiency distribution data of each waterfront green space in the target area, and then statistically obtain the average carbon sequestration efficiency of each waterfront green space.

[0075] S2. Construct an influencing factor index set, including two categories: urban waterfront green space form indexes and water-green space relationship indexes.

[0076] Among them, the influencing factor index set in S2 includes: two waterfront green space form indexes, namely canopy coverage rate and fractal dimension; five water-green space relationship indexes, namely water-green contact degree, water-green width ratio, proportion of forest land area within 10m, forest land within 10 - 20m, and proportion of forest land area within 20 - 30m.

[0077] S3. Based on the index set in S2 and the land use data in S1, calculate the data of waterfront green space form indexes and water-green space relationship indexes in the target area to obtain the influencing factor index data. The specific calculation steps are as follows:

[0078] S31. According to the land use data in S11 and the vector ranges of each waterfront green space in the target area in S16, in the ArcGIS software, use the "clip" tool to obtain the land use data of each waterfront green space. Subsequently, use the "raster to vector" tool to obtain the vector data of forest land and green space of each waterfront green space.

[0079] S32. According to the vector data of forest land and green space of each waterfront green space in S31, measure the forest land area and total green space area in different waterfront green spaces. The formula for calculating the canopy coverage rate of waterfront green space is:

[0080]

[0081] Among them, T i is the canopy coverage rate of the i-th waterfront green space, a i is the direct solar shadow area of the forest land canopy of the i-th waterfront green space, and b i is the total green space area of the i-th waterfront green space.

[0082] S33. According to the vector data of forest land and green space of each waterfront green space in S31, measure the total green space area and green space perimeter in different waterfront green spaces, and calculate the fractal dimension of the green space. The formula is:

[0083]

[0084] Among them, FRAC is the fractal dimension of the green space, p iis the perimeter of the i-th waterfront green space; a i is the area of the i-th waterfront green space.

[0085] S34. According to the forest land and green space vector data of each waterfront green space in S31, exported through ArcGIS software, measure the shoreline length of the waterfront green space using CAD software, and combine with the perimeter of each waterfront green space measured by ArcGIS software to calculate the water-green contact degree of the waterfront green space. The formula is:

[0086]

[0087] Among them, R i is the water-green contact degree of the i-th waterfront green space, L i is the shoreline length of the i-th waterfront green space, P i is the perimeter of the i-th waterfront green space.

[0088] S35. According to the land use data and remote sensing image data in S11, in ArcGIS software, determine the centerlines of the waterfront green space and the river through the binary method, and convert the centerlines of the river and the waterfront green space and the vector files of the waterfront green space and the river surface into CAD files; after opening in CAD software, determine the corresponding river channel range by making perpendicular lines according to the range of the waterfront green space and perform cropping. Subsequently, make the perpendicular bisectors of the centerlines of the waterfront green space and the corresponding river channel centerlines, and use the "path array" tool to perform array operations with the perpendicular bisector as the object and the centerlines of the waterfront green space and the corresponding river channel as the path, modify the array parameters, adjust to an appropriate spacing, crop the redundant part and then import it into ArcGIS software. After converting to vector data, obtain the average width of the waterfront green space and the corresponding river channel. The formula for calculating the water-green width ratio is:

[0089]

[0090] Among them, W i is the water-green width ratio of the i-th waterfront green space, W_R i is the average width of the corresponding river section of the i-th waterfront green space, W_G i is the average width of the i-th waterfront green space.

[0091] S36. According to the forest land and green space vector data of each waterfront green space in S31 and the land use data in S11, use the "buffer" tool in ArcGIS software to obtain each buffer of the water body, and then use the "clip" tool to obtain the forest land range of the waterfront green space within 10, 10 - 20, and 20 - 30 m from the shoreline, and calculate the proportion of the forest land area within 10, 10 - 20, and 20 - 30 m. The formula is:

[0092]

[0093] Among them, Pij is the ratio of the forest area of the i-th waterfront green space in different buffer zones to the total forest area of the i-th green space, A ij is the forest area of the i-th waterfront green space in different buffer zones. d i is the total forest area of the i-th waterfront green space. j are buffer zones at different distances from the shoreline, which are 10m, 10 - 20m, and 20 - 30m respectively.

[0094] S4. Based on the average carbon sequestration efficiency of waterfront green spaces obtained from S1 and the impact factor index data obtained from S3, construct an impact factor index data set, and calculate the importance of urban waterfront green space morphological indicators and water - green space relationship indicators through a random forest regression model. The top three or top five in importance are the key factors affecting the carbon sequestration benefits of urban waterfront green spaces. The results are as Figure 6 and the specific steps are as follows:

[0095] First, use the average carbon sequestration efficiency of the waterfront green spaces in the target area in S1 as the dependent variable, and the impact factor index data of the target area in S3 as the independent variable to construct a data set. Second, divide it into a training set and a test set in a ratio of 7:3, adopt 5 - fold cross - validation, with the maximum number of leaf nodes being 50, the maximum depth of the tree being 10, and the number of decision trees being 100 to construct a random forest model. Take R 2 , Mean Absolute Error as the necessary conditions for the model to pass accuracy verification.

[0096] Among them, the calculation formula of R 2 is:

[0097]

[0098] Among them, k is the number of folds of cross - validation, that is, the number of times of dividing the training set and the test set. The division ratio for constructing the response model is 0.8, and the k value is 5. y i is the predicted value of the dependent variable for the i - th calculation, is the actual value of the dependent variable for the i - th calculation, is the average value of the dependent variable, and n is the number of samples.

[0099] Second, the calculation formula of Mean Absolute Error is:

[0100]

[0101] Among them, k is the number of folds of cross - validation, y i is the predicted value of the dependent variable, is the actual value of the dependent variable, and n is the number of samples in the data set.

[0102] It should be noted that the above content only illustrates the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. For those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can still be made, and these improvements and refinements all fall within the protection scope of the claims of the present invention.

Claims

1. A method for identifying key influencing factors of the carbon sequestration benefit of urban waterfront green spaces, characterized in that, It includes the following steps: S1. Obtain the remote sensing image data, land use data, and NDVI data of the target area, and calculate the average carbon sequestration efficiency of each waterfront green space in the target area by combining the measured inversion of LAI and the measured investigation of trees and shrubs. S2. Construct an impact factor index set, including two categories: the morphological index of urban waterfront green spaces and the water-green space relationship index. S3. Based on the index set in step S2 and the land use data in step S1, calculate the morphological index data of the waterfront green spaces and the water-green space relationship index data in the target area to obtain the impact factor index data. S4. Based on the average carbon sequestration efficiency of the waterfront green spaces obtained in step S1 and the impact factor index data obtained in step S3, construct a regression analysis data set, and calculate the importance of the morphological index of urban waterfront green spaces and the water-green space relationship index through a random forest regression model. The top five in importance are the key factors affecting the carbon sequestration benefit of urban waterfront green spaces.

2. The method for identifying key influencing factors of carbon sequestration benefits of urban waterfront green spaces according to claim 1, characterized in that, In step S1, the specific steps for calculating the average carbon sequestration efficiency of each waterfront green space in the target area are as follows: S11. Obtain the remote sensing image data of the target area, and perform image supervised classification in the ArcGIS software to obtain the land use data of the target area. S12. Based on the remote sensing image data of the target area, calculate the NDVI data of the target area on the Envi platform. Among them, the calculation formula is: Among them, NIR and R respectively correspond to the near-infrared spectral and red spectral channels in the remote sensing image data. S13. Based on the land use data in step S11, randomly select 30 LAI measurement sample points and 200 tree and shrub investigation sample points in the green spaces of the target area through the ArcGIS software. Determine the quadrats with the geometric centers of the sample points. The shape of the quadrat is a square or approximate square of 10×10m. Measure the LAI values of trees and shrubs in the quadrats with a canopy analyzer to obtain the LAI values of the quadrats; record the types of trees and shrubs in the tree and shrub investigation quadrats to obtain the types of trees and shrubs in different quadrats. S14. Take the LAI value of the quadrat in step S13 as the dependent variable, and the NDVI value corresponding to the LAI measurement quadrat in step S13 as the independent variable to establish a linear regression equation. Combine the NDVI data in step S12 to invert the LAI value of the green spaces in the target area to obtain the LAI data of the green spaces in the target area. S15. Combine the types of trees and shrubs in the tree and shrub investigation quadrats in step S13, and take the types with higher frequencies as the common trees and shrubs in the green spaces of the target area. Based on the measured carbon sequestration efficiency data of each common tree and shrub, take its average value as the carbon sequestration efficiency of the trees and shrubs in the green spaces of the target area. Combine the LAI data of the green spaces in the target area in step S14 to calculate the carbon sequestration efficiency of the green spaces in the target area. Among them, the formula for calculating the carbon sequestration efficiency of the green spaces in the target area is: Among them, C is the daily carbon sequestration efficiency of the green space (g·m-2·d-1); y is the LAI value of the tree or shrub; c is the average daily carbon sequestration amount per unit leaf area of the i-th plant type (g·m-2·d-1); t is the effective photosynthetic days of the i-th plant type (d); i is different plant species. S16. Based on the remote sensing image data in step S11, in the ArcGIS software, using roads as boundaries, divide the waterfront green spaces within the target area to obtain the vector ranges of each waterfront green space in the target area. Combining with the carbon sequestration efficiency of the green spaces in the target area in S15, in the ArcGIS software, use the "Clip" tool to obtain the carbon sequestration efficiency distribution data of each waterfront green space in the target area, and then statistically obtain the average carbon sequestration efficiency of each waterfront green space.

3. The method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces according to claim 1, wherein The influencing factor index set in step S2 includes: two waterfront green space morphological indexes, namely crown cover rate and fractal dimension; five water-green space relationship indexes, namely water-green contact degree, water-green width ratio, proportion of forest land area within 10m, forest land from 10 - 20m, and proportion of forest land area from 20 - 30m.

4. A method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces according to claim 1, characterized in that The specific steps for calculating the influencing factor index data of each waterfront green space in the target area in step S3 are as follows: S31. According to the land use data in step S11 and the vector ranges of each waterfront green space in the target area in S16, in the ArcGIS software, use the "Clip" tool to obtain the land use data of each waterfront green space; subsequently, use the "Raster to Vector" tool to obtain the vector data of forest land and green space of each waterfront green space. S32. According to the vector data of forest land and green space of each waterfront green space in step S31, measure the forest land area and total green space area in different waterfront green spaces. The formula for calculating the crown cover rate of the waterfront green space is: Among them, T i is the crown density of the i-th waterfront green space, a i is the direct solar shadow area of the forest canopy in the i-th waterfront green space, b i is the total green area of the i-th waterfront green space; S33. According to the vector data of forest land and green space of each waterfront green space in step S31, measure the total green space area and green space perimeter in each waterfront green space, and calculate the fractal dimension of the green space. The formula is: Among them, FRAC is the fractal dimension of the green space, and p i is the perimeter of the i-th waterfront green space; a i is the area of the i-th waterfront green space; S34. According to the vector data of forest land and green space of each waterfront green space in step S31, export it through the ArcGIS software, use the CAD software to measure the shoreline length of each waterfront green space, and combine with the perimeter of each waterfront green space measured by the ArcGIS software to calculate the water-green contact degree of the waterfront green space. The formula is: Among them, R i is the water-green contact degree of the i-th waterfront green space, L i is the waterfront green space shoreline length of the i-th waterfront green space, P i is the perimeter of the i-th waterfront green space; S35. According to the land use data and remote sensing image data in step S11, in the ArcGIS software, use the binary method to determine the center lines of the waterfront green space and the river; import the waterfront green space, the river, and their center lines into the CAD software, draw the perpendicular bisectors of the center lines of the waterfront green space and the corresponding river center line, use the "Path Array" tool, with the perpendicular bisector as the object and the center lines of the waterfront green space and the corresponding river as the path for array operation, modify the array parameters, adjust to an appropriate spacing, crop and then import it into the ArcGIS software, and convert it into vector data to obtain the average width of the waterfront green space and the corresponding river; the formula for calculating the water-green width ratio is: Among them, \(W\) i is the water - green width ratio of the \(i\)th waterfront green space, \(W_R\) i is the average width of the river section corresponding to the \(i\)th waterfront green space, \(W_G\) i is the average width of the \(i\)th waterfront green space; S36. According to the vector data of forest land and green space of each waterfront green space in step S31 and the land use data in step S11, use the "Buffer" tool of the ArcGIS software to obtain each buffer zone of the water body, and then use the "Clip" tool to obtain the forest land ranges of the waterfront green space within 10m, 10 - 20m, and 20 - 30m from the shoreline, and calculate the proportion of forest land area within 10m, 10 - 20m, and 20 - 30m. The formula is: Among them, P ij is the ratio of the forest area of the i-th waterfront green space to the total forest area of the i-th green space in different buffer zones, and A ij is the forest area of the i-th waterfront green space in different buffer zones; d i is the total forest area of the i-th waterfront green space. The buffer zones at different distances from the shoreline are 10m, 10 - 20m, and 20 - 30m respectively.

5. A method for identifying key influencing factors of carbon sequestration benefits in urban waterfront green spaces according to claim 1, characterized in that In step S4, the specific steps for calculating the importance of each influencing factor of urban waterfront green space are as follows: First, taking the average carbon sequestration efficiency of the waterfront green space in the target area in step S1 as the dependent variable and the index data of the influencing factors in the target area in step S3 as the independent variable, a regression analysis data set is constructed; Second, it is divided into a training set and a test set in a ratio of 7:3, 5-fold cross-validation is adopted, the maximum number of leaf nodes is 50, the maximum depth of the tree is 10, and the number of decision trees is 100. A random forest model is constructed to analyze the regression analysis data set to obtain the importance of each factor. Take R 2 , the mean absolute error as a necessary condition for the model to pass the accuracy verification; Among them, R 2 The calculation formula is: Among them, k is the number of folds of cross-validation, that is, the number of times of dividing the training set and the test set. The division ratio for constructing the response model is 0.8, and the value of k is 5; y i is the predicted value of the dependent variable for the i-th calculation, is the actual value of the dependent variable for the i-th calculation, is the average value of the dependent variable, and n is the number of samples; Secondly, the calculation formula for the mean absolute error is: Among them, k is the number of folds of cross-validation, and y i is the predicted value of the dependent variable, is the actual value of the dependent variable, and n is the number of samples in the dataset.

Images