A multi-indicator-based method for assessing ecosystem service value of mixed bamboo and broadleaf forests

Abstract

The present invention relates to the technical field of ecological environment assessment, and discloses a method for assessing the service value of a bamboo-broadleaf mixed forest ecosystem based on multiple indicators. The method comprises the following steps: step S101, collecting remote sensing data of the bamboo-broadleaf mixed forest; step S102, generating a feature vector; step S103, constructing spatiotemporal graph structure data; step S104, measuring carbon flux by using a carbon flux measurement model; step S105, measuring soil erosion by using a soil erosion measurement model; and step S106, calculating and obtaining a temperature adjustment factor and a humidity adjustment factor of the bamboo-broadleaf mixed forest. The method constructs remote sensing data of multiple sub-areas of the bamboo-broadleaf mixed forest into spatiotemporal graph structure data, performs spatiotemporal analysis on the spatiotemporal graph structure data by using the carbon flux measurement model, establishes a nonlinear mapping relationship between the remote sensing data and the carbon flux, thereby achieving large-area carbon flux measurement, and further provides a temperature adjustment factor and a humidity adjustment factor to implement a multi-indicator assessment method for the bamboo-broadleaf mixed forest.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological environment assessment, and more specifically, to a method for assessing the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators. Background Art

[0002] Mixed bamboo and broad-leaved forests are a type of forest ecosystem composed of bamboo and broad-leaved tree species. They play an important role in providing a variety of ecosystem services, including carbon sequestration, soil and water conservation, and biodiversity protection. However, existing methods for assessing the value of ecosystem services in mixed bamboo and broad-leaved forests have the following defects: 1. Data collection methods are limited, mainly relying on ground surveys and laboratory analysis, which makes it difficult to cover large areas and is time-consuming and labor-intensive; 2. Data analysis methods are not perfect, lacking advanced spatial analysis and dynamic monitoring technologies, making it difficult to reflect the dynamic changes and long-term development trends of ecosystems; 3. There is a lack of a unified indicator system, and usually only focuses on one-sided service functions, such as only focusing on carbon storage (including vegetation carbon storage and soil carbon storage), but carbon flux is also one of the important indicators.

[0003] Therefore, a multi-indicator-based ecosystem service value assessment method for mixed bamboo and broad-leaved forests is urgently needed to solve the above problems. Summary of the Invention

[0004] The present invention provides a multi-indicator-based method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem, which solves the technical problems in the above-mentioned background technology.

[0005] The present invention provides a method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators, comprising the following steps:

[0006] Step S101, dividing the bamboo and broadleaf mixed forest into M sub-areas, and collecting remote sensing data of each sub-area at a preset time interval t within a preset time period T;

[0007] The number of sub-areas M, the preset time period T, and the preset time interval t are all custom parameters;

[0008] Step S102, pre-processing the remote sensing data of each sub-region to generate a feature vector of uniform size;

[0009] The eigenvector includes eight dimension values, corresponding to the normalized difference vegetation index, enhanced vegetation index, soil-adjusted vegetation index, surface temperature, photosynthetically active radiation, evapotranspiration, soil moisture, and leaf area index;

[0010] Step S103, constructing spatiotemporal graph structure data based on the feature vectors of the M sub-regions;

[0011] The spatiotemporal graph structure data includes N graph structure data, where N=T / t;

[0012] Each graph structure data includes M nodes, and the mth node establishes a mapping relationship with the mth sub-region, where 1≤m≤M;

[0013] The node feature of the mth node of the nth graph structure data is represented by the feature vector of the mth sub-region at the nth time point with which the mapping relationship is established, where 1≤n≤N;

[0014] The edges between nodes in the spatiotemporal graph structure data include: edges constructed between adjacent nodes of the same graph structure data; edges constructed between the same nodes of adjacent graph structure data;

[0015] Step S104, inputting the spatiotemporal graph structure data into a carbon flux measurement model, and outputting a value representing the carbon flux of the bamboo-broadleaved mixed forest;

[0016] Step S105, inputting the spatiotemporal graph structure data into a soil erosion amount determination model, and outputting a value representing the soil erosion amount of the bamboo-broadleaved mixed forest;

[0017] Step S106 , obtaining the surface temperature and humidity of the ground without vegetation cover and the surface temperature and humidity of the ground with vegetation cover in the M sub-areas at N time points, and respectively calculating the temperature adjustment factor and humidity adjustment factor of the bamboo and broad-leaved mixed forest.

[0018] Furthermore, the normalized vegetation index of a pixel is calculated based on the reflectivity corresponding to the near-infrared band and the red band of the remote sensing data of the sub-region, and the average value of the normalized vegetation index of all pixels of the remote sensing data of the sub-region is calculated as the normalized vegetation index of the sub-region; according to the above steps, the enhanced vegetation index of the sub-region is calculated based on the reflectivity corresponding to the near-infrared band, the blue band and the red band of the remote sensing data; according to the above steps, the soil adjusted vegetation index of the sub-region is calculated based on the reflectivity corresponding to the near-infrared band and the red band of the remote sensing data.

[0019] Furthermore, the surface temperature of the sub-region is obtained by calculating the blackbody radiation formula based on the reflectivity corresponding to the thermal infrared band of the remote sensing data of the sub-region; the photosynthetically active radiation of the sub-region is obtained by inversion calculation based on the reflectivity corresponding to the shortwave infrared band and blue light band of the remote sensing data of the sub-region; the evapotranspiration is obtained by inversion calculation based on the surface temperature of the sub-region and the reflectivity corresponding to the visible light band and near-infrared band of the remote sensing data; the soil moisture is obtained by inversion calculation based on the normalized vegetation index of the sub-region; and the leaf area index is obtained by inversion calculation based on the enhanced vegetation index of the sub-region.

[0020] Furthermore, the carbon flux determination model includes N first hidden layers, 1 second hidden layer and 1 classifier;

[0021] The nth first hidden layer inputs the nth graph structure data of the spatiotemporal graph network and outputs a feature matrix, where 1≤n≤N;

[0022] The feature matrix consists of M row vectors, each row vector corresponds to the update vector of a node;

[0023] The feature matrix output by the Nth first hidden layer is input to the second hidden layer, and the second hidden layer outputs the result vector;

[0024] The result vector output by the second hidden layer is input into the classifier, and the classification space of the classifier represents the carbon flux of the bamboo and broad-leaved mixed forest.

[0025] Furthermore, the calculation formula of the carbon flux determination model includes:

[0026] The calculation formula for the nth first hidden layer includes:

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] in represents the feature matrix of the nth first hidden layer output, and Represents the update vector of the mth node of the feature matrix of the nth and n-1th first hidden layer outputs, respectively, Assigned to , Represents the set of nodes that have an edge connection with the mth node of the nth graph structure data. and Respectively represent the node features of the mth and gth nodes of the nth graph structure data, and Respectively represent the first intermediate vector and the second intermediate vector of the mth node of the nth graph structure data, 、 、 and They represent the first weight parameter, the second weight parameter, the third weight parameter and the fourth weight parameter of the nth first hidden layer respectively, and They represent the first bias parameter and the second bias parameter of the nth first hidden layer respectively, Represents the gating coefficient of the first hidden layer of the nth layer. The gating coefficient is a real number ranging from 0 to 1. Indicates stacking the update vectors of M nodes, concat indicates concatenation, Swish indicates the Swish activation function, and sigmoid indicates the sigmoid activation function.

[0033] The calculation formula for the second hidden layer is as follows:

[0034] ;

[0035] in represents the result vector output by the second hidden layer, The feature matrix representing the Nth first hidden layer output of the second hidden layer input, and denote the first weight parameter and the second weight parameter of the second hidden layer respectively, b denotes the bias parameter of the second hidden layer, and tanh denotes the hyperbolic tangent activation function.

[0036] Furthermore, the calculation formula for the sample label of the training sample used to train the carbon flux determination model is as follows:

[0037] ;

[0038] Where NEE represents carbon flux, Indicates the air density, which is measured by a gas density meter. Represents the specific heat capacity of air, which is measured by an air specific heat capacity meter. Represents the molecular diffusion coefficient, which is calculated using the existing empirical formula. represents the respiratory temperature response coefficient, It is a custom parameter with a value range of 1.5 to 3.0. Indicates the current temperature. Indicates the reference temperature, which is assigned a value of 25°C, Carbon indicates the carbon dioxide concentration, and Wind indicates the vertical wind speed. represents the covariance of carbon dioxide concentration and vertical wind speed, represents the standard deviation of vertical wind speed, represents the correction coefficient, and f represents the air friction speed.

[0039] Furthermore, the structure and calculation formula of the soil erosion determination model are the same as those of the carbon flux determination model.

[0040] Furthermore, the sample labels of the training samples used to train the soil erosion measurement model are calculated by using the universal soil loss equation.

[0041] Furthermore, the temperature regulation factor of mixed bamboo and broadleaf forests The calculation formula is as follows:

[0042] ;

[0043] in and They represent the surface temperature without vegetation cover and the surface temperature with vegetation cover at the nth time point in the mth sub-area respectively;

[0044] Humidity regulating factors in mixed bamboo and broad-leaved forests The calculation formula is as follows:

[0045] ;

[0046] in and They represent the surface moisture without vegetation cover and the surface moisture with vegetation cover at the nth time point in the mth sub-area, respectively.

[0047] Furthermore, the method further comprises the following steps:

[0048] Step S107, obtaining a comprehensive score based on the weighted sum of the carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor of the mixed bamboo and broadleaf forest;

[0049] The weight coefficients corresponding to carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor are all custom parameters whose sum is 1.

[0050] The beneficial effects of the present invention are as follows: the present invention constructs the remote sensing data of multiple sub-areas of the bamboo-broadleaved mixed forest into spatiotemporal graph structure data, performs spatiotemporal analysis on the spatiotemporal graph structure data through a carbon flux measurement model, and establishes a nonlinear mapping relationship between the remote sensing data and the carbon flux, thereby realizing large-area carbon flux measurement. In addition, according to the concept of the present invention, large-area soil erosion and carbon storage measurement can also be realized, and the present invention also provides a temperature adjustment factor and a humidity adjustment factor to realize a multi-indicator evaluation method for the bamboo-broadleaved mixed forest. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention is a flowchart of a method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators. DETAILED DESCRIPTION

[0052] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0053] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprising" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but do not exclude other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0054] It should be noted that carbon flux refers to the amount of carbon dioxide absorbed or released by an ecosystem per unit time, which is expressed by NEE (net ecosystem exchange). NEE represents the net absorption (positive value) or net release (negative value) of carbon dioxide by the ecosystem per unit time. It is the difference between GPP (gross primary productivity) and Re (ecosystem respiration), where GPP represents the total amount of carbon dioxide fixed by the ecosystem through photosynthesis per unit time, and Re represents the total amount of carbon dioxide released into the atmosphere by the ecosystem per unit time.

[0055] Existing carbon flux measurements usually use eddy covariance instruments installed on observation towers to measure the vertical gradient changes of meteorological elements such as wind speed, temperature, humidity, and CO2 concentration in real time, and obtain NEE through relevant calculation formulas. However, the eddy covariance method relies on wind-driven vortex transport to capture carbon flux signals. Under low-flux conditions (such as at night or low wind speed), vortex activity weakens, resulting in a deviation between the measured carbon flux and the actual carbon flux.

[0056] Therefore, the present invention combines remote sensing data with a carbon flux measurement model to perform spatiotemporal analysis, establishes a nonlinear mapping relationship between remote sensing data and carbon flux, and realizes large-area carbon flux measurement.

[0057] like Figure 1 As shown in FIG, a multi-indicator-based method for assessing the ecosystem service value of mixed bamboo and broadleaf forests includes the following steps:

[0058] Step S101, dividing the bamboo and broadleaf mixed forest into M sub-areas, and collecting remote sensing data of each sub-area at a preset time interval t within a preset time period T;

[0059] Step S102, pre-processing the remote sensing data of each sub-region to generate a feature vector of uniform size;

[0060] The eigenvector includes eight dimension values, corresponding to the normalized difference vegetation index, enhanced vegetation index, soil-adjusted vegetation index, surface temperature, photosynthetically active radiation, evapotranspiration, soil moisture, and leaf area index;

[0061] Step S103, constructing spatiotemporal graph structure data based on the feature vectors of the M sub-regions;

[0062] The spatiotemporal graph structure data includes N graph structure data, where N=T / t;

[0063] Each graph structure data includes M nodes, and the mth node establishes a mapping relationship with the mth sub-region, where 1≤m≤M;

[0064] The node feature of the mth node of the nth graph structure data is represented by the feature vector of the mth sub-region at the nth time point with which the mapping relationship is established, where 1≤n≤N;

[0065] The edges between nodes in the spatiotemporal graph structure data include: edges constructed between adjacent nodes of the same graph structure data; edges constructed between the same nodes of adjacent graph structure data;

[0066] Step S104, inputting the spatiotemporal graph structure data into a carbon flux measurement model, and outputting a value representing the carbon flux of the bamboo-broadleaved mixed forest;

[0067] Step S105, inputting the spatiotemporal graph structure data into a soil erosion amount determination model, and outputting a value representing the soil erosion amount of the bamboo-broadleaved mixed forest;

[0068] Step S106 , obtaining the surface temperature and humidity of the ground without vegetation cover and the surface temperature and humidity of the ground with vegetation cover in the M sub-areas at N time points, and respectively calculating the temperature adjustment factor and humidity adjustment factor of the bamboo and broad-leaved mixed forest.

[0069] It should be noted that remote sensing data is represented by pixels, and each pixel is represented by the reflectance corresponding to the visible light band, near infrared band (760nm~900nm), short wave infrared band (1570nm~1750nm and 2080nm~2350nm) and thermal infrared band (8500nm~9300nm and 10400nm~12500nm). The visible light band includes: blue light band (450nm~520nm), green light band (520nm~600nm) and red light band (6 30nm~690nm), it can be seen that the amount of remote sensing data is huge, which is not conducive to subsequent data analysis and processing. Therefore, the remote sensing data of each sub-region is extracted by preprocessing, which retains the characteristics of the remote sensing data while greatly reducing the data dimension, thereby reducing the computational complexity of the carbon flux determination model and improving the calculation speed. In addition, the remote sensing data of each sub-region are unified into feature vectors of the same size by preprocessing, which can also facilitate the spatiotemporal analysis and calculation of the carbon flux determination model, thereby improving the calculation speed.

[0070] In one embodiment of the present invention, the number of sub-areas M, the preset time period T and the preset time interval t are all custom parameters. Preferably, M is set to 20, the preset time period T is set to 10 days, and the preset time interval t is set to 1 day. Remote sensing data can be uniformly collected at 10 am or 2 pm every day to avoid radiation differences caused by changes in the solar altitude angle. In addition, the atmospheric correction model can be used to eliminate the influence of atmospheric absorption and scattering on remote sensing data. The atmospheric correction model can be MODTRAN, 6S model, etc., which will not be described in detail here.

[0071] In one embodiment of the present invention, a normalized vegetation index of a pixel is calculated based on the reflectivity corresponding to the near-infrared band and the red light band of the remote sensing data of the sub-region, and the average value of the normalized vegetation indexes of all pixels of the remote sensing data of the sub-region is calculated as the normalized vegetation index of the sub-region; according to the above steps, an enhanced vegetation index of the sub-region is calculated based on the reflectivity corresponding to the near-infrared band, the blue light band and the red light band of the remote sensing data; according to the above steps, a soil-adjusted vegetation index of the sub-region is calculated based on the reflectivity corresponding to the near-infrared band and the red light band of the remote sensing data.

[0072] According to the above embodiment, the normalized vegetation index of the mth sub-region is The calculation formula is as follows:

[0073] ;

[0074] Where 1≤m≤M, K represents the number of pixels of remote sensing data in the sub-region, and They represent the reflectance corresponding to the near-infrared band and the red band of the k-th pixel of the remote sensing data of the m-th sub-region respectively.

[0075] According to the above embodiment, the enhanced vegetation index of the mth sub-region is The calculation formula is as follows:

[0076] ;

[0077] in It represents the reflectance of the blue light band of the kth pixel of the remote sensing data of the mth sub-region. G represents the gain factor, which is assigned a value of 2.5. Indicates the red light coefficient, assigned a value of 6, Indicates the blue light coefficient, the value is 7.5, Represents the atmospheric correction constant, which is assigned a value of 1.

[0078] According to the above embodiment, the soil adjusted vegetation index of the mth sub-region is The calculation formula is as follows:

[0079] ;

[0080] in Represents the soil brightness factor, with a value of 0.5.

[0081] In one embodiment of the present invention, the surface temperature of the sub-region is obtained by calculating the blackbody radiation formula based on the reflectivity corresponding to the thermal infrared band of the remote sensing data of the sub-region; the photosynthetically active radiation of the sub-region is obtained by inversion calculation based on the reflectivity corresponding to the short-wave infrared band and the blue light band of the remote sensing data of the sub-region; the evapotranspiration is obtained by inversion calculation based on the surface temperature of the sub-region and the reflectivity corresponding to the visible light band and the near-infrared band of the remote sensing data; the soil moisture is obtained by inversion calculation based on the normalized vegetation index of the sub-region; and the leaf area index is obtained by inversion calculation based on the enhanced vegetation index of the sub-region.

[0082] It should be noted that the inversion calculation can fit the relationship between the sub-region data and the measured data through the least squares method or other regression methods. For example, the calculation formula of the leaf area index is as follows: leaf area index = weight coefficient × enhanced vegetation index + bias coefficient, which will not be elaborated here.

[0083] In one embodiment of the present invention, the carbon flux determination model includes N first hidden layers, 1 second hidden layer and 1 classifier;

[0084] The nth first hidden layer inputs the nth graph structure data of the spatiotemporal graph network and outputs a feature matrix, where 1≤n≤N;

[0085] The feature matrix consists of M row vectors, each row vector corresponds to the update vector of a node;

[0086] The feature matrix output by the Nth first hidden layer is input to the second hidden layer, and the second hidden layer outputs the result vector;

[0087] The result vector output by the second hidden layer is input into the classifier, and the classification space of the classifier represents the carbon flux of the bamboo and broad-leaved mixed forest.

[0088] In one embodiment of the present invention, the calculation formula of the carbon flux determination model includes:

[0089] The calculation formula for the nth first hidden layer includes:

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] in represents the feature matrix of the nth first hidden layer output, and Represents the update vector of the mth node of the feature matrix of the nth and n-1th first hidden layer outputs, respectively, Assigned to , Represents the set of nodes that have an edge connection with the mth node of the nth graph structure data. and Respectively represent the node features of the mth and gth nodes of the nth graph structure data, and Respectively represent the first intermediate vector and the second intermediate vector of the mth node of the nth graph structure data, 、 、 and They represent the first weight parameter, the second weight parameter, the third weight parameter and the fourth weight parameter of the nth first hidden layer respectively, and They represent the first bias parameter and the second bias parameter of the nth first hidden layer respectively, Represents the gating coefficient of the first hidden layer of the nth layer. The gating coefficient is a real number ranging from 0 to 1. Indicates stacking the update vectors of M nodes, concat indicates concatenation, Swish indicates the Swish activation function, and sigmoid indicates the sigmoid activation function.

[0096] The calculation formula for the second hidden layer is as follows:

[0097] ;

[0098] in represents the result vector output by the second hidden layer, The feature matrix representing the Nth first hidden layer output of the second hidden layer input, and denote the first weight parameter and the second weight parameter of the second hidden layer respectively, b denotes the bias parameter of the second hidden layer, and tanh denotes the hyperbolic tangent activation function.

[0099] It should be noted that the weight parameters and bias parameters in the carbon flux determination model are all learnable hyperparameters. For example, if the size of the feature matrix is ​​M×8, the size of the node update vector is 1×8. The node update vector and the node feature (feature vector) of the node are concatenated to obtain a vector of size 1×16. It can be designed as a 16×1 vector, and the two are multiplied and then activated by the sigmoid function to obtain the gate coefficient. and It can be designed as a matrix of size 8×16, then is a 1×16 vector, is a 1×8 vector, so the two can be concatenated to get a 1×24 vector. Designed as a 24×8 matrix, then according to the above example, It can be designed as a vector of size 1×M. It can be designed as a matrix of size 8×16, then the result vector output by the second hidden layer is a vector of size 1×16, and the activation function of the classifier is the softmax activation function, which will not be described here.

[0100] In one embodiment of the present invention, the calculation formula for the sample label of the training sample used to train the carbon flux determination model is as follows:

[0101] ;

[0102] Where NEE represents carbon flux, Indicates the air density, which is measured by a gas density meter. Represents the specific heat capacity of air, which is measured by an air specific heat capacity meter. Represents the molecular diffusion coefficient, which is calculated using existing empirical formulas, such as the Fuller molecular diffusion coefficient empirical formula. represents the respiratory temperature response coefficient, It is a custom parameter with a value range of 1.5 to 3.0. Indicates the current temperature. Indicates the reference temperature, which is assigned a value of 25°C, Carbon indicates the carbon dioxide concentration, and Wind indicates the vertical wind speed. represents the covariance of carbon dioxide concentration and vertical wind speed, represents the standard deviation of vertical wind speed, represents the correction coefficient, and f represents the air friction speed.

[0103] It should be noted that air density, temperature and wind speed will affect the measurement of carbon flux. Therefore, the present invention comprehensively considers the influence of air density, temperature and wind speed on the basis of the eddy covariance method, thereby improving the accuracy of carbon flux measurement. The correction coefficient can be obtained by fitting the experimental data, which will not be elaborated here.

[0104] In one embodiment of the present invention, the structure and calculation formula of the soil erosion measurement model are the same as those of the carbon flux measurement model, and are not described in detail here.

[0105] In one embodiment of the present invention, the sample labels of the training samples used to train the soil erosion measurement model are obtained by calculation using the USLE (Universal Soil Loss Equation) or the RUSLE (Revised Universal Soil Loss Equation).

[0106] In one embodiment of the present invention, the temperature regulation factor of the bamboo and broadleaf mixed forest is The calculation formula is as follows:

[0107] ;

[0108] in and They represent the surface temperature without vegetation cover and the surface temperature with vegetation cover at the nth time point in the mth sub-area respectively;

[0109] Humidity regulating factors in mixed bamboo and broad-leaved forests The calculation formula is as follows:

[0110] ;

[0111] in and They represent the surface moisture without vegetation cover and the surface moisture with vegetation cover at the nth time point in the mth sub-area, respectively.

[0112] In one embodiment of the present invention, a multi-indicator-based method for assessing the service value of a mixed bamboo and broadleaf forest ecosystem provided by the present invention further includes the following steps:

[0113] Step S107, obtaining a comprehensive score based on the weighted sum of the carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor of the mixed bamboo and broadleaf forest;

[0114] The weight coefficients corresponding to carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor are all custom parameters whose sum is 1. Preferably, the weight coefficients of carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor are set to 0.4, 0.4, 0.1, and 0.1, respectively.

[0115] It should be noted that, according to the ideas provided by the present invention, large-scale carbon storage measurement can also be achieved, such as vegetation carbon storage measurement models and soil carbon storage measurement models. Similarly, by constructing spatiotemporal graph structure data as sample data for training vegetation carbon storage measurement models and soil carbon storage measurement model training samples, sample labels for training vegetation carbon storage measurement models and soil carbon storage measurement model training samples are obtained through field sampling and experimental analysis. In addition, species richness can be obtained through field surveys and added to the calculation of comprehensive scores. Relevant experts can set thresholds to determine whether the ecological environment of the bamboo and broad-leaved mixed forest meets the standards, or set thresholds for the carbon flux, carbon storage, soil erosion, temperature regulation factor, humidity regulation factor and species richness of the bamboo and broad-leaved mixed forest to determine whether the requirements are met, etc., which will not be elaborated here.

[0116] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A multi-index-based method for assessing the ecosystem service value of mixed bamboo and broadleaf forests, characterized in that: The following steps are involved: Step S101, dividing the bamboo and broadleaf mixed forest into M sub-areas, and collecting remote sensing data of each sub-area at a preset time interval t within a preset time period T; The number of sub-areas M, the preset time period T, and the preset time interval t are all custom parameters; Step S102, pre-processing the remote sensing data of each sub-region to generate a feature vector of uniform size; The eigenvector includes eight dimension values, corresponding to the normalized difference vegetation index, enhanced vegetation index, soil-adjusted vegetation index, surface temperature, photosynthetically active radiation, evapotranspiration, soil moisture, and leaf area index; Step S103, constructing spatiotemporal graph structure data based on the feature vectors of the M sub-regions; The spatiotemporal graph structure data includes N graph structure data, where N=T / t; Each graph structure data includes M nodes, and the mth node establishes a mapping relationship with the mth sub-region, where 1≤m≤M; The node feature of the mth node of the nth graph structure data is represented by the feature vector of the mth sub-region at the nth time point with which the mapping relationship is established, where 1≤n≤N; The edges between nodes in the spatiotemporal graph structure data include: edges constructed between adjacent nodes of the same graph structure data; edges constructed between the same nodes of adjacent graph structure data; Step S104, inputting the spatiotemporal graph structure data into a carbon flux measurement model, and outputting a value representing the carbon flux of the bamboo-broadleaved mixed forest; Step S105, inputting the spatiotemporal graph structure data into a soil erosion amount determination model, and outputting a value representing the soil erosion amount of the bamboo-broadleaved mixed forest; Step S106 , obtaining the surface temperature and humidity of the ground without vegetation cover and the surface temperature and humidity of the ground with vegetation cover in the M sub-areas at N time points, and respectively calculating the temperature adjustment factor and humidity adjustment factor of the bamboo and broad-leaved mixed forest.

2. The method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, characterized in that: The normalized vegetation index of a pixel is calculated based on the reflectance corresponding to the near-infrared band and the red band of the remote sensing data of the sub-region, and the average value of the normalized vegetation index of all pixels of the remote sensing data of the sub-region is calculated as the normalized vegetation index of the sub-region; according to the above steps, the enhanced vegetation index of the sub-region is calculated based on the reflectance corresponding to the near-infrared band, the blue band and the red band of the remote sensing data; according to the above steps, the soil adjusted vegetation index of the sub-region is calculated based on the reflectance corresponding to the near-infrared band and the red band of the remote sensing data.

3. The method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, characterized in that: The surface temperature of the sub-region is obtained by calculating the blackbody radiation formula based on the reflectivity corresponding to the thermal infrared band of the remote sensing data of the sub-region; the photosynthetically active radiation of the sub-region is obtained by inversion calculation based on the reflectivity corresponding to the shortwave infrared band and blue light band of the remote sensing data of the sub-region; the evapotranspiration is obtained by inversion calculation based on the surface temperature of the sub-region and the reflectivity corresponding to the visible light band and near-infrared band of the remote sensing data; the soil moisture is obtained by inversion calculation based on the normalized vegetation index of the sub-region; and the leaf area index is obtained by inversion calculation based on the enhanced vegetation index of the sub-region.

4. The method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, wherein: The carbon flux determination model includes N first hidden layers, 1 second hidden layer and 1 classifier; The nth first hidden layer inputs the nth graph structure data of the spatiotemporal graph network and outputs a feature matrix, where 1≤n≤N; The feature matrix consists of M row vectors, each row vector corresponds to the update vector of a node; The feature matrix output by the Nth first hidden layer is input to the second hidden layer, and the second hidden layer outputs the result vector; The result vector output by the second hidden layer is input into the classifier, and the classification space of the classifier represents the carbon flux of the bamboo and broad-leaved mixed forest.

5. The method for evaluating the service value of mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 4, characterized in that: The calculation formula of the carbon flux determination model includes: The calculation formula for the nth first hidden layer includes: ； ； ； ； ； in represents the feature matrix of the nth first hidden layer output, and Represents the update vector of the mth node of the feature matrix of the nth and n-1th first hidden layer outputs, respectively, Assigned to , Represents the set of nodes that have an edge connection with the mth node of the nth graph structure data. and Respectively represent the node features of the mth and gth nodes of the nth graph structure data, and Respectively represent the first intermediate vector and the second intermediate vector of the mth node of the nth graph structure data, 、 、 and They represent the first weight parameter, the second weight parameter, the third weight parameter and the fourth weight parameter of the nth first hidden layer respectively, and They represent the first bias parameter and the second bias parameter of the nth first hidden layer respectively, Represents the gating coefficient of the first hidden layer of the nth layer. The gating coefficient is a real number ranging from 0 to 1. Indicates stacking the update vectors of M nodes, concat indicates concatenation, Swish indicates Swish activation function, and sigmoid indicates sigmoid activation function; The calculation formula for the second hidden layer is as follows: ； in represents the result vector output by the second hidden layer, The feature matrix representing the Nth first hidden layer output of the second hidden layer input, and denote the first weight parameter and the second weight parameter of the second hidden layer respectively, b denotes the bias parameter of the second hidden layer, and tanh denotes the hyperbolic tangent activation function.

6. The method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, characterized in that: The calculation formula for the sample label of the training sample used to train the carbon flux determination model is as follows: ； Where NEE represents carbon flux, Indicates the air density, which is measured by a gas density meter. Represents the specific heat capacity of air, which is measured by an air specific heat capacity meter. Represents the molecular diffusion coefficient, which is calculated using the existing empirical formula. represents the respiratory temperature response coefficient, It is a custom parameter with a value range of 1.5 to 3.

0. Indicates the current temperature. Indicates the reference temperature, which is assigned a value of 25°C, Carbon indicates the carbon dioxide concentration, and Wind indicates the vertical wind speed. represents the covariance of carbon dioxide concentration and vertical wind speed, represents the standard deviation of vertical wind speed, represents the correction coefficient, and f represents the air friction speed.

7. The method for evaluating the service value of mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 5, characterized in that: The structure and calculation formula of the soil erosion determination model are the same as those of the carbon flux determination model.

8. The method for evaluating the service value of mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, characterized in that: The sample labels of the training samples used to train the soil erosion determination model are calculated using the universal soil loss equation.

9. The method for evaluating the service value of a mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, wherein: Temperature regulating factors in mixed bamboo and broad-leaved forests The calculation formula is as follows: ； in and They represent the surface temperature without vegetation cover and the surface temperature with vegetation cover at the nth time point in the mth sub-area respectively; Humidity regulating factors in mixed bamboo and broad-leaved forests The calculation formula is as follows: ； in and They represent the surface moisture without vegetation cover and the surface moisture with vegetation cover at the nth time point in the mth sub-area, respectively.

10. The method for evaluating the service value of mixed bamboo and broadleaf forest ecosystem based on multiple indicators according to claim 1, characterized in that: The following steps are also included: Step S107, obtaining a comprehensive score based on the weighted sum of the carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor of the mixed bamboo and broadleaf forest; The weight coefficients corresponding to carbon flux, soil erosion, temperature adjustment factor, and humidity adjustment factor are all custom parameters whose sum is 1.

Images