Ecological benefit prediction method for urban blue-green space

Abstract

Disclosed in the present invention is an ecological benefit prediction method for an urban blue-green space, comprising the following steps: S1, establishing an urban blue-green space morphological feature index set and an ecological benefit index set; S2, calculating urban blue-green space morphological feature indexes, ecological benefit indexes, and a total ecological benefit, calculating the correlation and importance among the morphological feature indexes, and screening for key morphological feature indexes; S3, using a random forest algorithm to train key morphological feature index data and total ecological benefit data to obtain a response relationship between morphology and ecology; S4, calculating morphological index data after implementation of an urban blue-green space planning scheme; and S5, on the basis of the response relationship between morphology and ecology, predicting ecological benefits after the implementation of the urban blue-green space planning scheme by means of the morphological index data after the implementation of the urban blue-green space planning scheme. In the present invention, a random forest machine learning algorithm is used to predict the ecological benefits after the implementation of the urban blue-green space planning scheme, thereby providing a reference for ecological environment benefit evaluation of urban development.

Description

Urban blue-green space ecological benefit prediction method TECHNICAL FIELD

[0001] The present application belongs to the field of urban ecological environment benefit evaluation, and particularly relates to an urban blue-green space ecological benefit prediction method. BACKGROUND

[0002] With the city entering the stock development stage, the blue-green space is the main body of ecological benefit in high-density cities and is an important object of attention for current high-quality urban development. The morphological characteristics of urban blue-green space are closely related to ecological benefits and influence each other, while in the past urban construction, the spatial form and ecological benefits are separated. The existing research on ecological benefit prediction mostly focuses on forests or urban green spaces, and lacks ecological benefit prediction methods after the implementation of urban planning schemes. In addition, the previous ecological benefit prediction focuses on single green space, ignoring the synergistic effect of blue-green space as a whole, which not only cannot accurately reflect the correlation between form and ecology, but also is affected by human factors, interest factors and cognitive limitations, and the prediction accuracy is not high.

[0003] SUMMARY

[0004] To solve the above problems, the present application discloses an urban blue-green space ecological benefit prediction method, which adopts a random forest machine learning algorithm to realize ecological benefit prediction of urban blue-green space planning schemes, thereby providing a reference for ecological environment benefit evaluation of urban development.

[0005] To achieve the above purpose, the technical scheme of the present application is as follows:

[0006] An urban blue-green space ecological benefit prediction method, specifically comprising the following steps:

[0007] S1, establishing an urban blue-green space morphological characteristic index set and an ecological benefit index set;

[0008] S2, calculating urban blue-green space morphological characteristic indexes, ecological benefit indexes and total ecological benefits, calculating the correlation and importance between morphological characteristic indexes, and screening key morphological characteristic indexes;

[0009] S3, based on the key morphological characteristic index data and the total ecological benefit data, a random forest algorithm is used to train the response relationship between form and ecology;

[0010] S4, calculating and obtaining key morphological index data of urban blue-green space planning schemes;

[0011] S5, based on the response relationship between form and ecology, the ecological benefit of the urban blue-green space planning scheme is predicted through the morphological index data of the urban blue-green space planning scheme.

[0012] Preferably, the urban blue-green space form feature index set S1 comprises two-dimensional form feature indexes and three-dimensional form feature indexes;

[0013] The two-dimensional form feature indexes comprise: urban green space average perimeter-area ratio PARA-MN 绿 , urban green space average fractal dimension FRAC-MN 绿 , urban green space landscape shape index LSI 绿 , urban green space aggregation index AI 绿 , urban green space cohesion COHESION 绿 , urban green space connectivity CONNECT 绿 , urban green space separation DIVISION 绿 , urban green space splitting SPLIT 绿 , urban green space patch number NP 绿 , urban green space patch density PD 绿 , urban blue space average perimeter-area ratio PARA-MN 蓝 , urban blue space average fractal dimension FRAC-MN 蓝 , urban blue space landscape shape index LSI 蓝 , urban blue space aggregation index AI 蓝 , urban blue space cohesion COHESION 蓝 , urban blue space connectivity CONNECT 蓝 , urban blue space separation DIVISION 蓝 , urban blue space splitting SPLIT 蓝 , urban blue space patch number NP 蓝 , and urban blue space patch density PD 蓝 .

[0014] The three-dimensional form feature indexes comprise: urban green space total three-dimensional area VCA 绿 , urban green space maximum patch three-dimensional area proportion VLPI 绿 , urban green space core patch three-dimensional area ratio VCPLAND 绿 , urban blue space total three-dimensional area VCA 蓝 , urban blue space maximum patch three-dimensional area proportion VLPI 蓝 , urban blue space core patch three-dimensional area ratio VCPLAND 蓝 .

[0015] The urban blue-green space ecological benefit index set comprises urban surface temperature LST, net primary productivity NPP, carbon storage C total , surface runoff Q, and waterlogging point density Fn and habitat quality HQ;

[0016] Preferably, the calculation of the urban blue-green space form feature index, the ecological benefit index and the total ecological benefit specific process of S2 comprises:

[0017] Step 1, obtaining the basic data of the target area, wherein the basic data comprises remote sensing data, Digital Elevation Model (DEM) data, urban land cover data, urban soil type data and meteorological monitoring station data.

[0018] Step 2, using Fragstats software platform type layer landscape pattern index to calculate the form feature index;

[0019] Step 3, based on remote sensing data, using the classic Radiative Transfer Equation (RTE) to calculate the urban surface temperature LST; based on the Carnegie-Ames-Stanford Approach (CASA) model to calculate the net primary productivity NPP; based on the Carbon Storage and Sequestration module of the Integrated Valuation of Ecosystem Services and Trade-offs (InVEST) platform to calculate the carbon storage C total ; based on the Soil Conservation Service-Curve Number (SCS-CN) model to calculate the surface runoff Q; based on the remote sensing image analysis method to identify the waterlogging area and calculate the waterlogging point density Fn; based on the Habitat Quality module of the InVEST platform to calculate the habitat quality HQ.

[0020] Step 4, using the average weight method to calculate the total ecological benefit. The calculation of the total ecological benefit comprises: the ecological benefit is mainly composed of cooling benefit, carbon sink benefit, rainwater resilience benefit and biodiversity protection benefit; the urban surface temperature LST indicates the cooling benefit, the net primary productivity NPP and the carbon storage C total indicate the carbon sink benefit, the surface runoff Q and the waterlogging point density Fn indicate the rainwater resilience benefit, and the habitat quality HQ indicates the biodiversity protection benefit.

[0021] The average weight method is used to calculate the total ecological benefit. First, the positive index and the negative index are normalized and standardized.

[0022] The positive index comprises the net primary productivity NPP and the carbon storage Ctotal and habitat quality HQ, the calculation formula of positive indicators is as follows:

[0023] Wherein, X i ′ is the normalized value of the i th ecological benefit index, X i is the original value of the i th ecological benefit index, X imin is the minimum value of the i th ecological benefit index, and X imax is the maximum value of the i th ecological benefit index;

[0024] Negative indicators include urban land surface temperature LST, surface runoff Q and waterlogging point density Fn, and the calculation formula of negative indicators is as follows:

[0025] Based on the normalization and standardization of positive and negative indicators, the weights of the four types of ecological benefit indicators are considered as the same value and are averaged. Among them, the carbon sink benefit and rainwater resilience benefit are averaged after the two indicators are averaged, and then the total ecological benefit (ECO) is calculated.

[0026] The calculation formula of total ecological benefit (ECO) is as follows:

[0027] ECO = 0.25LST′ + 0.25(0.5NPP′ + 0.5C total ′)

[0028] + 0.25(0.5Q′ + 0.5F n ′) + 0.25HQ′

[0029] Preferably, the calculation of waterlogging point density specifically includes:

[0030] Step 1, based on remote sensing image analysis method to identify waterlogging area, the calculation formula is as follows:

[0031] Wherein, i represents the land use of the study area on sunny days, j represents the land use of the study area after continuous rainy days, n is the number of land use types, S ij is the conversion of land use types before and after rainfall; according to the calculation result of S ij , the area where the land use type is converted from non-water body to water body after rainfall is regarded as waterlogging area;

[0032] Step 2, convert the waterlogging area into vector points and calculate the waterlogging point density, the calculation formula is as follows:

[0033] Wherein, F n is the waterlogging point density, N is the number of waterlogging points in the area, and S is the area of the region.

[0034] Preferably, the specific process of screening the key morphological characteristic indicators in S2 comprises the following steps:

[0035] Step 1, using SPSS software (Statistical Product and Service Solutions, also known as social science statistical software package) to calculate the correlation between morphological indicators;

[0036] Step 2, using out-of-bag data to calculate the importance of morphological characteristic indicators;

[0037] Step 3, screening out the indicators with less importance and the indicators with high correlation and less importance, to obtain the key morphological characteristic indicators;

[0038] Preferably, the specific process of training the response relationship between morphology and ecology by using the random forest algorithm in S3 comprises the following steps:

[0039] Step 1, based on the key morphological characteristic indicator data and the total ecological benefit data, constructing a training set and a test set;

[0040] Step 2, setting the random forest operation parameters in the response relationship, wherein the operation parameters include the number of decision trees, the maximum depth of the decision tree, the minimum leaf tree and the random seed number;

[0041] Step 3, using the training set to train the response relationship, and using the test set to verify the accuracy of the response relationship, to obtain the optimized response relationship between morphology and ecology.

[0042] The beneficial effects of the present application are:

[0043] 1, the present application breaks through the precise prediction of the ecological benefits of blue-green space, and only needs to be based on the morphological characteristics of the planning scheme to quickly evaluate the ecological benefits, and fills the blank in the past prediction field.

[0044] 2, compared with the time-consuming and labor-consuming of the traditional prediction method, the present application greatly improves the efficiency by introducing the random forest algorithm, and enhances the accuracy of the prediction results, and provides strong technical support for ecological planning and management. BRIEF DESCRIPTION OF DRAWINGS

[0045] Fig. 1 is a flowchart of the present application;

[0046] Fig. 2 is the urban blue-green space ecological benefit chart of the present application;

[0047] Fig. 3 is the center city waterlogging point density chart of the present application;

[0048] Fig. 4 is the morphological indicator correlation coefficient chart of the present application;

[0049] Figure 5 is a contrastive diagram of the importance of the morphological indicators according to the present application;

[0050] Figure 6 is a scatter plot of actual value-predicted value in the precision verification of the random forest algorithm according to the present application;

[0051] Figure 7 is a distribution map of the blue-green space in the central urban area of Nanjing in 2035 according to an embodiment of the present application;

[0052] Figure 8 (a-e) is the calculated morphological characteristics of the blue-green space in the central urban area of Nanjing in 2035 according to an embodiment of the present application, which are VCA 蓝 , VLPI 绿 , AI 绿 , LSI 绿 , and COHESION 蓝 , respectively;

[0053] Figure 9 is a map of the ecological benefits of the central urban area of Nanjing in 2035 according to an embodiment of the present application. DETAILED DESCRIPTION

[0054] The present application will be further illustrated below in conjunction with the drawings and specific embodiments, and it should be understood that the following specific embodiments are only used to illustrate the present application and not to limit the scope of the present application.

[0055] Taking Nanjing as an example, a method for predicting the ecological benefits of urban blue-green space according to the present application, as shown in Figure 1, includes the following steps:

[0056] S1, establishing a set of morphological characteristic indicators and a set of ecological benefit indicators of urban blue-green space;

[0057] The set of morphological characteristic indicators of urban blue-green space includes two-dimensional morphological indicators and three-dimensional morphological indicators;

[0058] The two-dimensional morphological indicators include: urban green space average perimeter area ratio PARA-MN 绿 , urban green space average fractal dimension FRAC-MN 绿 , urban green space landscape shape index LSI 绿 , urban green space aggregation index AI 绿 , urban green space sprawl index COHESION 绿 , urban green space connectivity CONNECT 绿 , urban green space separation DIVISION 绿 , urban green space splitting SPLIT 绿 , urban green space patch number NP 绿 , urban green space patch density PD 绿 , urban blue space average perimeter area ratio PARA-MN 蓝, average fractal dimension of urban blue space FRAC-MN 蓝 , landscape shape index of urban blue space LSI 蓝 , aggregation index of urban blue space AI 蓝 , cohesion of urban blue space COHESION 蓝 , connectivity of urban blue space CONNECT 蓝 , division of urban blue space DIVISION 蓝 , split of urban blue space SPLIT 蓝 , number of patches of urban blue space NP 蓝 , and patch density of urban blue space PD 蓝 ;

[0059] The three-dimensional morphological indicators include: total three-dimensional area of urban green space VCA 绿 , three-dimensional area proportion of the largest patch of urban green space VLPI 绿 , three-dimensional area proportion of the core patch of urban green space VCPLAND 绿 , total three-dimensional area of urban blue space VCA 蓝 , three-dimensional area proportion of the largest patch of urban blue space VLPI 蓝 , three-dimensional area proportion of the core patch of urban blue space VCPLAND 蓝 .

[0060] The urban blue-green space ecological benefit indicator set includes urban surface temperature LST, net primary productivity NPP, carbon storage C total , surface runoff Q, waterlogging point density Fn, and habitat quality HQ.

[0061] S2, calculating urban blue-green space morphological characteristic indicators, ecological benefit indicators and total ecological benefits, and screening key morphological characteristic indicators;

[0062] The specific process of calculating urban blue-green space morphological characteristic indicators, ecological benefit indicators and total ecological benefits includes:

[0063] Step 1, obtaining basic data of the research area, wherein the basic data includes remote sensing data, DEM data, urban land cover data, urban soil type data and meteorological monitoring station data.

[0064] Specifically, in this embodiment, the multispectral remote sensing data is derived from the Landsat-8 satellite remote sensing image set (Path-Row: 120-38) disclosed on the official website of the United States Geological Survey (https: / / www.usgs.gov / ). It is used for the identification of urban land types, the inversion of land surface temperature, the simulation of carbon sink data, and the identification of waterlogging areas. The DEM data is derived from the SRTM v3.1 (Shuttle Radar Topography Mission) data set measured and collected by the United States Space Agency (NASA, https: / / www.earthdata.nasa.gov / ), which is used for the calculation of three-dimensional morphological indicators. The urban land cover data (GLC_FCS30-2020) is derived from the 30-meter resolution fine feature data set disclosed by the China Aerospace Institute (http: / / data.casearth.cn / sdo / detail), which is used for simulating carbon sink data, calculating carbon storage, simulating runoff, and calculating habitat quality. The urban soil type data is derived from the China Soil Database (Soil Science Database, http: / / vdb3.soil.csdb.cn / ), which includes soil type distribution, soil moisture content, and other data, and is used for simulating runoff. In addition, Nanjing meteorological data is also needed when simulating carbon sink data and simulating runoff, mainly using Nanjing meteorological monitoring station data, and obtaining through Kriging interpolation calculation.

[0065] Step 2, calculate morphological characteristic indicators using Fragstats software platform type layer landscape pattern index;

[0066] Step 3, based on remote sensing data, calculate urban surface temperature LST using the classic transfer equation (Radiative Transfer Equation, RTE); calculate net primary productivity NPP based on the CASA model; calculate carbon storage C based on the Carbon Storage and Sequestration module of the InVEST platform total ; calculate surface runoff Q based on the SCS-CN model; identify waterlogging areas based on remote sensing image analysis method, calculate waterlogging point density Fn; calculate habitat quality HQ based on the Habitat Quality module of the InVEST platform.

[0067] Step 4, calculate the total ecological benefit by the average weight method. The calculation of the total ecological benefit includes: the ecological benefit is mainly composed of cooling benefit, carbon sink benefit, rainwater resilience benefit and biodiversity protection benefit, urban surface temperature LST indicates cooling benefit, net primary productivity NPP and carbon storage C total indicate carbon sink benefit, surface runoff Q, waterlogging point density F nThe rain flood resilience benefit and the habitat quality HQ indicate the biodiversity protection benefit.

[0068] The total ecological benefit is calculated using the average weight method, and the calculation result is shown in FIG. 2. Each ecological benefit index is divided into a positive index and a negative index. First, the positive index and the negative index are normalized and standardized.

[0069] The positive index includes net primary productivity NPP, carbon storage C total and habitat quality HQ, and the calculation formula of the positive index is as follows:

[0070] wherein X i ′ is the normalized value of the i th ecological benefit index, X i is the original value of the i th ecological benefit index, X imin is the minimum value of the i th ecological benefit index, and X imax is the maximum value of the i th ecological benefit index;

[0071] The negative index includes urban surface temperature LST, surface runoff Q and waterlogging point density F n , and the calculation formula of the negative index is as follows:

[0072] Based on the normalized and standardized results of the positive and negative indexes, the weights of the four types of ecological benefit indexes are considered as the same value and are averaged. The carbon sink benefit and the rain flood resilience benefit are averaged after the two indexes involved are averaged, and then the total ecological benefit (ECO) is calculated again.

[0073] The calculation formula of the total ecological benefit (ECO) is as follows:

[0074] ECO = 0.25LST′ + 0.25(0.5NPP′ + 0.5C total ′)

[0075] + 0.25(0.5Q′ + 0.5F n ′) + 0.25HQ′

[0076] Further, the calculation of the waterlogging point density specifically includes:

[0077] Step 1, waterlogging area identification based on remote sensing image analysis method, and the calculation formula is as follows:

[0078] wherein i represents the land use condition of the research area on a sunny day, j represents the land use condition of the research area after continuous rainy days, n is the number of land use types, S ij is the conversion of land use types before and after the rain; and S ijThe calculation result takes the area where the land use of the non-water body before the rainfall is converted into a water body after the rainfall as the waterlogging area;

[0079] Step 2, convert the waterlogging area into a vector point, and calculate the waterlogging point density, and the result is shown in FIG. 3, and the calculation formula is as follows:

[0080] Wherein, F n is the waterlogging point density, N is the number of waterlogging points in the area, and S is the area of the area.

[0081] Further, the specific process of screening the key morphological characteristic index includes:

[0082] Step 1, calculate the correlation between morphological indexes on the SPSS platform, as shown in FIG. 4;

[0083] Step 2, calculate the importance of morphological characteristic indexes by using the out-of-bag data estimation method, as shown in FIG. 5;

[0084] Step 3, first, the indexes with less importance are screened out, second, the indexes with high correlation and less importance are screened out, and finally, the key morphological characteristic indexes are determined; the final key indexes are: VCA 蓝 , VLPI 绿 , AI 绿 , LSI 绿 , COHESION 蓝 .

[0085] S3, based on the key morphological characteristic index data and the total ecological benefit data, a response relationship between morphology and ecology is trained by using a random forest algorithm;

[0086] Step 1, based on the key morphological characteristic index data and the total ecological benefit data, a training set and a test set are constructed; in this embodiment, 80% of the data are randomly selected as the training set, and the remaining 20% of the data are used as the test set.

[0087] Step 2, set the random forest operation parameters in the response relationship, and the operation parameters include the number of decision trees, the maximum depth of the decision tree, the minimum leaf tree, and the random seed number; in this embodiment, the number of decision trees is 100, the maximum depth of the decision tree is 10, the minimum leaf tree is 8, and the random seed number is 43.

[0088] Step 3, the training set is used to train the response relationship, and the test set is used to verify the accuracy of the response relationship, and an optimized response relationship between morphology and ecology is obtained. As shown in FIG. 6.

[0089] S4, calculate the obtained urban blue-green space planning scheme key form index data; in this embodiment, the overall planning scheme of Nanjing city in the overall planning of Nanjing city (2021-2035) is selected, the overall planning scheme of the central urban area of Nanjing is extracted as the basic data, and the blue-green space data is calculated, as shown in FIG. 7; the key form characteristic indexes are calculated by using ArcGIS (Arc Geographic Information System) software and Fragstats software, as shown in FIG. 8.

[0090] S5, based on the response relationship between the form and the ecology, the ecological benefits of the urban blue-green space planning scheme are obtained by using the form index data of the urban blue-green space planning scheme.

[0091] The calculated form characteristic indexes are run in the trained random forest algorithm, and the corresponding planning scheme ecological benefits are output, as shown in FIG. 9.

[0092] It should be noted that the above content only illustrates the technical idea of the present application, and cannot be used to limit the protection scope of the present application. For ordinary skilled persons in the technical field, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements fall within the protection scope of the claims of the present application.

Claims

1. A method for predicting ecological benefits of urban blue-green spaces, characterized in that, The method comprises the following steps: S1, establishing a set of urban blue-green space form feature indexes and a set of ecological benefit indexes; S2, calculating the urban blue-green space form feature indexes, the ecological benefit indexes and the total ecological benefit, calculating the correlation and importance between the form feature indexes, and screening the key form feature indexes; S3, based on the key form feature index data and the total ecological benefit data, using a random forest algorithm to train and obtain the response relationship between the form and the ecology; S4, calculating the key form index data of the obtained urban blue-green space planning scheme; S5, based on the response relationship between the form and the ecology, predicting the ecological benefit of the urban blue-green space planning scheme through the form index data of the urban blue-green space planning scheme. 2.The urban blue-green space ecological benefit prediction method according to claim 1, characterized in that, The set of urban blue-green space form feature indexes in step S1 comprises two-dimensional form feature indexes and three-dimensional form feature indexes; The two-dimensional morphological characteristic index includes: urban green space average perimeter area ratio PARA-MN 绿 , urban green space average fractal dimension FRAC-MN 绿 , urban green space landscape shape index LSI 绿 , urban green space aggregation degree AI 绿 , urban green space spread COHESION 绿 , urban green space connectivity CONNECT 绿 , urban green space separation DIVISION 绿 , urban green space splitting SPLIT 绿 , urban green space patch number NP 绿 , urban green space patch density PD 绿 , urban blue space average perimeter area ratio PARA-MN 蓝 , urban blue space average fractal dimension FRAC-MN 蓝 , urban blue space landscape shape index LSI 蓝 , urban blue space aggregation degree AI 蓝 , urban blue space spread COHESION 蓝 , urban blue space connectivity CONNECT 蓝 , urban blue space separation DIVISION 蓝 , urban blue space splitting SPLIT 蓝 , urban blue space patch number NP 蓝 , and urban blue space patch density PD 蓝 ; The three-dimensional morphological characteristic index comprises: total three-dimensional area of urban green space VCA 绿 , three-dimensional area proportion of the largest patch of urban green space VLPI 绿 , three-dimensional area ratio of the core patch of urban green space VCPLAND 绿 , total three-dimensional area of urban blue space VCA 蓝 , three-dimensional area proportion of the largest patch of urban blue space VLPI 蓝 , three-dimensional area ratio of the core patch of urban blue space VCPLAND 蓝 ; The urban blue-green space ecological benefit index set includes urban surface temperature LST, net primary productivity NPP, carbon storage C total , surface runoff Q, waterlogging point density Fn, and habitat quality HQ. 3.The urban blue-green space ecological benefit prediction method of claim 1, wherein, The specific process of calculating the urban blue-green space form feature indexes, the ecological benefit indexes and the total ecological benefit in step S2 comprises: Step 1, obtaining the basic data of the target region, wherein the basic data comprises remote sensing data, DEM data, urban land cover data, urban soil type data and meteorological monitoring station data; Step 2, using the Fragstats software platform type layer landscape pattern index to calculate the form feature indexes; Step 3, based on remote sensing data, using the classic transfer equation to calculate the urban surface temperature LST; based on the CASA model to calculate the net primary productivity NPP; based on the Carbon Storage and Sequestration module of the InVEST platform to calculate the carbon storage C total ; based on the SCS-CN model to calculate the surface runoff Q; based on the remote sensing image analysis method to identify the waterlogging area and calculate the waterlogging point density Fn; based on the Habitat Quality module of the InVEST platform to calculate the habitat quality HQ; Step 4, calculating total ecological benefits by using the average weight method; the calculation of total ecological benefits includes: ecological benefits mainly consist of cooling benefits, carbon sink benefits, rain flood resilience benefits and biodiversity protection benefits, urban surface temperature LST indicates cooling benefits, net primary productivity NPP and carbon storage C total indicate carbon sink benefits, surface runoff Q and waterlogging point density Fn indicate rain flood resilience benefits, and habitat quality HQ indicates biodiversity protection benefits.

4. The method of claim 3, wherein, The total ecological benefit is calculated using the average weight method, and the specific process is as follows: Each ecological benefit index is divided into a positive index and a negative index; first, the positive index and the negative index are normalized and standardized; Positive indicators include net primary productivity (NPP), carbon storage (C), and habitat quality (HQ), which are calculated as follows: total and habitat quality (HQ), which are calculated as follows: wherein X i is the normalized value of the i-th ecological benefit index, X i is the original value of the i-th ecological benefit index, X imin is the minimum value of the i-th ecological benefit index, and X imax is the maximum value of the i-th ecological benefit index; The negative indicators include urban surface temperature LST, surface runoff Q, and waterlogging point density Fn, and the calculation formulas of the negative indicators are as follows: Based on the normalized and standardized results of the positive and negative indexes, the weights of the four types of ecological benefit indexes are considered as the same numerical average weight; wherein the carbon sink benefit and the rainwater resilience benefit are calculated again after being averaged and weighted, and then the total ecological benefit is calculated; The calculation formula of the total ecological benefit is as follows: ECO = 0.25 LST' + 0.25 (0.5 NPP' + 0.5 C total ') + 0.25 (0.5 Q' + 0.5F n ' ) + 0.25 H Q '.

5. The method of claim 3, wherein, The specific content of calculating the density of waterlogging points comprises: Step 1, the waterlogging area is identified based on a remote sensing image analysis method, and a calculation formula is as follows: where i represents the land use of the study area on a sunny day, j represents the land use of the study area after consecutive rainy days, n is the number of land types, S ij is the conversion of land types before and after rainfall; according to S ij The calculation result is that the area where the land type before rainfall is converted into water body after rainfall is regarded as the waterlogging area. Step 2, the waterlogging area is converted into a vector point, and the waterlogging point density is calculated, and the calculation formula is as follows: where F n is the density of waterlogging points, N is the number of waterlogging points in the region, and S is the area of the region. 6.The urban blue-green space ecological benefit prediction method according to claim 1, characterized in that, The specific process of screening the key form feature indexes in step S2 comprises: Step 1, using the SPSS software to calculate the correlation between the form feature indexes; Step 2, using the out-of-bag data to calculate the importance of the form feature indexes; Step 3, removing the indexes with small importance and the indexes with high correlation and small importance to obtain the key form feature indexes.

7. The method of claim 1, wherein, The specific process of using the random forest algorithm to train and obtain the response relationship between the form and the ecology in step S3 comprises: Step 1, based on the key form feature index data and the total ecological benefit data, constructing a training set and a test set; Step 2, setting the random forest operation parameters in the response relationship, wherein the operation parameters comprise the number of decision trees, the maximum depth of the decision tree, the minimum leaf tree and the random seed number; Step 3, using the training set to train the response relationship, and using the test set to verify the accuracy of the response relationship, to obtain the optimized response relationship between the form and the ecology.

Images