Carbon dynamic acquisition method and long-term sustainability evaluation method for forestry carbon sink project

Abstract

The invention discloses a carbon dynamic acquisition method and a long-term sustainability evaluation method for a forestry carbon sink project. The carbon dynamic acquisition method comprises the following steps: acquiring multi-source data of a target forestry carbon sink project, and preprocessing the multi-source data to obtain a first data set; the multi-source data comprises ecological data, climate data, economic data and social data; constructing a multi-dimensional grid model based on the first data set; obtaining a carbon reserve polynomial fitting function based on the multi-dimensional grid model and a preset strategy; performing interpolation in the whole network of the multi-dimensional grid model by using a carbon reserve polynomial fitting function to obtain a polynomial model; and obtaining carbon dynamic information of the target forestry carbon sink project based on the polynomial model. According to the method, the actual condition of forestry carbon sink can be comprehensively reflected, high-precision carbon reserve prediction of different areas can be carried out, the obtained carbon dynamic information is comprehensive, reliable and accurate, and powerful support can be provided for dynamic monitoring and management of forestry carbon sink.

Description

Technical Field

[0001] The present disclosure relates to the technical field of carbon sink monitoring, and in particular to a carbon dynamic acquisition method and a long-term sustainability assessment method for a forestry carbon sink project. Background Art

[0002] As one of the important measures to address climate change, the long-term sustainability assessment of forestry carbon sink projects is crucial to the successful implementation and management of the projects. However, the methods disclosed in the prior art for long-term sustainability assessment have many limitations.

[0003] Traditional assessment methods often only focus on analyzing data in a single dimension, and it is difficult to comprehensively consider data information from multiple aspects such as ecology, economy and society. At the same time, these methods are also limited by the temporal and spatial scales and cannot flexibly respond to data changes at different temporal and spatial scales, resulting in poor prediction accuracy and unreliable results.

[0004] In addition, traditional assessment methods often have low computational efficiency and high resource consumption when processing large-scale, complex and multi-dimensional forestry carbon sink data. More importantly, these methods have weak dynamic adjustment capabilities and are difficult to make flexible adaptive adjustments based on real-time data and feedback, which greatly reduces the timeliness and accuracy of the assessment results. Summary of the invention

[0005] In view of this, the embodiments of the present disclosure provide a carbon dynamic acquisition method and a long-term sustainability assessment method for a forestry carbon sink project, which can solve the problems existing in the prior art such as unreliable prediction results, poor accuracy, poor timeliness, long preset cycles, and inability to perform adaptive adjustments based on real-time data and feedback.

[0006] In a first aspect, the present disclosure provides a method for dynamically acquiring carbon in a forestry carbon sink project, specifically comprising:

[0007] Collect multi-source data of the target forestry carbon sink project, and pre-process the multi-source data to obtain a first data set; the multi-source data includes ecological data, climate data, economic data and social data;

[0008] constructing a multidimensional grid model based on the first data set;

[0009] Obtaining a carbon storage polynomial fitting function based on the multidimensional grid model and a preset strategy;

[0010] interpolating the carbon storage polynomial fitting function within the entire network of the multidimensional grid model to obtain a polynomial model;

[0011] The carbon dynamic information of the target forestry carbon sink project is obtained based on the polynomial model.

[0012] Optionally, constructing a multidimensional grid model based on the first data set includes: determining a grid dimension of the target forestry carbon sink project based on the first data set; the grid dimension includes a time dimension, a space dimension, an ecological dimension, an economic dimension, and a social dimension;

[0013] Determining a grid resolution of each type of grid dimension; the grid resolution includes temporal resolution, spatial resolution, ecological resolution, economic resolution or social resolution;

[0014] Based on the grid dimension and the grid resolution, corresponding data points in the first data set are allocated to specific positions of the grid according to preset rules to generate the multi-dimensional grid model.

[0015] Optionally, the obtaining of the carbon storage polynomial fitting function based on the multidimensional grid model and a preset strategy includes: determining a target polynomial basis function based on the multidimensional grid model;

[0016] Obtain polynomial coefficients based on the least squares method;

[0017] A carbon storage polynomial fitting function is obtained based on the polynomial coefficients and the target polynomial basis function.

[0018] Optionally, the obtaining of polynomial coefficients based on the least squares method includes: constructing a polynomial matching model based on the target polynomial basis function, wherein the polynomial matching model includes a plurality of polynomial functions;

[0019] determining a least squares objective based on the polynomial matching model;

[0020] converting the polynomial matching model into a set of thread equations;

[0021] The thread equation group is solved based on the least squares objective to obtain polynomial coefficients.

[0022] Optionally, the obtaining of carbon dynamic information of the target forestry carbon sink project based on the polynomial model includes: constructing a partial differential equation model of the project based on the polynomial model;

[0023] The partial differential equation model of the project includes a group of dynamic partial differential equations of carbon storage, a group of dynamic partial differential equations of biomass and a group of dynamic partial differential equations of soil organic matter;

[0024] The partial derivatives of the partial differential equation model of the project are approximated based on the finite difference strategy at the grid points of the multidimensional grid model to obtain the rate of change of carbon storage.

[0025] Optionally, approximating partial derivatives of the project partial differential equation model based on a finite difference strategy at grid points of the multidimensional grid model to obtain the rate of change of carbon storage includes: using a finite difference strategy at grid points of the multidimensional grid model to approximate time derivatives and space derivatives;

[0026] Obtaining a target partial differential equation group based on the project partial differential equation model;

[0027] Determine the target basis function;

[0028] Based on the time derivative, the space derivative, the target partial differential equation group, the target basis function, and the Galerkin method, a discrete equation is constructed:

[0029] The boundary conditions are: the normal derivative on the boundary is 0;

[0030] Calculating the carbon storage at the next time step at each grid point based on the discrete equation, and obtaining the total carbon storage at each time step;

[0031] The total carbon storage at each time step is processed based on the time series to obtain the dynamic change information of the total carbon storage over time.

[0032] Optionally, also include:

[0033] Obtain historical climate characteristic data, historical economic characteristic data, historical social characteristic data and target ecological characteristic data of the target forestry carbon sink project;

[0034] The target ecological characteristic data include: current carbon storage and annual carbon absorption obtained based on the polynomial model, and normalized vegetation index extracted from remote sensing data;

[0035] A number of key features are selected from each type of feature data to obtain a key data set of the corresponding type;

[0036] Determine a number of initial learning models, the number of the initial learning models being consistent with the number of types of feature data;

[0037] Based on the initial learning model corresponding to each type of the key data set training, the trained initial learning model is verified by using a cross-validation strategy, and the parameters of the initial learning model that meets the verification requirements are tuned based on a Bayesian optimization strategy, and the tuned initial learning model is the target learning model;

[0038] Based on the first data set and the plurality of target learning models, obtaining prediction results of corresponding types;

[0039] A weighted average integration strategy is used to process a number of the prediction results to obtain target results, which include carbon storage long-term trend prediction information, project economic feasibility prediction information and ecosystem service value prediction information.

[0040] In a second aspect, the embodiment of the present disclosure further provides a method for evaluating the long-term sustainability of a forestry carbon sink project. The method is based on the carbon dynamic acquisition method of the forestry carbon sink project, and further includes:

[0041] Based on the carbon dynamic information, the annual carbon absorption and cumulative carbon storage within a future preset period are obtained, and a time series prediction chart is created;

[0042] Determine a sensitivity analysis heat map of the economic feasibility of the project based on the time series forecast graph;

[0043] Identify the key influencing factors in the target forestry carbon sink project, and obtain the project's long-term sustainability assessment information based on the key influencing factors and the sensitivity analysis heat map.

[0044] In a third aspect, the present disclosure also provides a carbon dynamic acquisition system for a forestry carbon sink project, including:

[0045] A collection module, used to collect multi-source data of a target forestry carbon sink project, and pre-process the multi-source data to obtain a first data set; the multi-source data includes ecological data, climate data, economic data, and social data;

[0046] A construction module, used for constructing a multidimensional grid model based on the first data set;

[0047] A fitting function acquisition module, used to obtain a carbon storage polynomial fitting function based on the multidimensional grid model and a preset strategy;

[0048] An interpolation module, used to interpolate within the entire network of the multidimensional grid model using the carbon storage polynomial fitting function to obtain a polynomial model;

[0049] The analysis module is used to obtain the carbon dynamic information of the target forestry carbon sink project based on the polynomial model.

[0050] In a fourth aspect, the embodiments of the present disclosure further provide a computer device, which adopts the following technical solution:

[0051] The computer device comprises:

[0052] at least one processor; and,

[0053] a memory communicatively connected to the at least one processor; wherein,

[0054] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the above-mentioned carbon dynamic acquisition methods for forestry carbon sink projects or the above-mentioned long-term sustainability assessment methods for forestry carbon sink projects.

[0055] In a fifth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium storing computer instructions for enabling a computer to execute any of the above-mentioned methods for dynamically acquiring carbon for a forestry carbon sink project or the above-mentioned methods for assessing the long-term sustainability of a forestry carbon sink project.

[0056] In a sixth aspect, an embodiment of the present disclosure further provides a computer program product, comprising a computer program / instruction, which implements the steps of any of the above methods when executed by a processor.

[0057] The carbon dynamics acquisition method for forestry carbon sink projects disclosed in the present application can comprehensively reflect the actual situation of forestry carbon sinks by integrating data from multiple sources including ecology, climate, economy, and society; a multidimensional grid model is constructed based on the preprocessed data, which is not limited to a fixed time and space scale and can adapt to different time and space scales; interpolation is performed using a polynomial fitting function to fill in the gaps in the data, making the carbon reserve information more comprehensive and can be used for carbon reserve prediction in different regions, that is, it can provide high-precision carbon reserve prediction and effectively reduce the uncertainty of the evaluation results. The detailed carbon dynamics information obtained can provide strong support for the dynamic monitoring and management of forestry carbon sinks, and can help formulate scientific forestry management and carbon emission reduction strategies and optimize resource allocation; in addition, the multidimensional grid model and the polynomial model can be adjusted according to different data and environmental conditions and have good adaptability; the method disclosed in this embodiment can effectively improve the comprehensiveness, accuracy and timeliness of the evaluation.

[0058] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1A schematic flow chart of a method for dynamically acquiring carbon in a forestry carbon sink project provided in an embodiment of the present disclosure.

[0061] Figure 2 A schematic flow chart of a method for constructing a multidimensional grid model provided in an embodiment of the present disclosure.

[0062] Figure 3 A schematic flow chart of a method for obtaining a carbon storage polynomial fitting function provided in an embodiment of the present disclosure.

[0063] Figure 4 A flowchart of a method for obtaining polynomial coefficients provided in an embodiment of the present disclosure.

[0064] Figure 5 A flow chart of a method for obtaining carbon dynamic information of a target forestry carbon sink project based on a polynomial model provided in an embodiment of the present disclosure.

[0065] Figure 6 A schematic flow chart of a method for obtaining the rate of change of carbon reserves provided in an embodiment of the present disclosure.

[0066] Figure 7 A schematic flow chart of a method for calculating the carbon storage of the next time step at each grid point based on a discrete equation provided in an embodiment of the present disclosure.

[0067] Figure 8 A flow chart of another embodiment of the method for dynamically acquiring carbon in a forestry carbon sink project provided in the embodiment of the present disclosure.

[0068] Fig. 9 A schematic flow chart of a method for assessing the long-term sustainability of a forestry carbon sink project provided in an embodiment of the present disclosure.

[0069] Fig.10 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0070] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0071] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0072] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0073] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0074] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0075] Reference Figure 1 The present application discloses a method for dynamically acquiring carbon in a forestry carbon sink project, which specifically includes the following schemes:

[0076] S100, collecting multi-source data of the target forestry carbon sink project, and pre-processing the multi-source data to obtain a first data set; the multi-source data includes ecological data, climate data, economic data and social data.

[0077] Specifically, ecological data include forest biomass, species diversity, soil carbon storage, forest land type, tree species distribution, biomass, soil type, etc.; climate data include meteorological data such as temperature, precipitation, humidity, wind speed, etc.; economic data include economic factors such as project investment, carbon credit income, timber output, forestry management costs, timber prices, land use changes, etc.; social data include employment opportunities, community participation, livelihood improvement, local community dependence on forestry, social policies, etc.

[0078] Preprocessing specifically includes one or more of outlier detection and processing, missing value interpolation, and data format unification.

[0079] This step can ensure the accuracy and consistency of the data, lay the foundation for subsequent analysis, and provide comprehensive basic data.

[0080] S200, constructing a multidimensional grid model based on the first data set.

[0081] Specifically, various types of data in the first data set are mapped into a multidimensional space, where each dimension corresponds to a data type (such as climate, ecology, etc.); a grid model is created, in which each grid cell represents a spatial area, and the dimensions of the grid reflect various aspects of the data.

[0082] In this step, the multidimensional grid model can intuitively represent complex environmental data for easy analysis, while providing a structured framework for efficient processing and analysis of large-scale data sets.

[0083] S300, obtaining a carbon storage polynomial fitting function based on a multidimensional grid model and a preset strategy.

[0084] In this step, the polynomial fitting function can capture the nonlinear relationship in the data, improve the prediction accuracy, and make the prediction of carbon storage more universal and adaptable.

[0085] S400, using the carbon stock polynomial fitting function to interpolate within the entire network of the multidimensional grid model to obtain a polynomial model.

[0086] Specifically, the obtained carbon stock polynomial fitting function is used in the multidimensional grid model to interpolate the unobserved areas within the grid and estimate the carbon stock in these areas, which can generate a complete polynomial model covering the carbon stock information of the entire grid network.

[0087] S500, obtains carbon dynamic information of target forestry carbon sink projects based on polynomial model.

[0088] Specifically, a polynomial model is used to calculate changes in carbon stocks at different time points or under different management measures, and a carbon dynamics report can be generated, including a review of historical data and a forecast of future trends.

[0089] Through this step, changes in carbon storage can be tracked and predicted in real time, providing decision support for forestry management, helping to evaluate the impact of different policies or management measures on carbon storage, and improving the scientific nature of carbon sink management.

[0090] The carbon dynamics information obtained can be used to calculate and analyze changes in total carbon storage over time, compare differences in carbon storage between different tree species and in different locations, and evaluate the effectiveness of forest carbon sink projects.

[0091] The carbon dynamics acquisition method for forestry carbon sink projects disclosed in the present application can comprehensively reflect the actual situation of forestry carbon sinks by integrating data from multiple sources including ecology, climate, economy, and society; a multidimensional grid model is constructed based on the preprocessed data, which is not limited to a fixed time and space scale and can adapt to different time and space scales; interpolation is performed using a polynomial fitting function to fill in the gaps in the data, making the carbon reserve information more comprehensive and can be used for carbon reserve prediction in different regions, that is, it can provide high-precision carbon reserve prediction and effectively reduce the uncertainty of the evaluation results. The detailed carbon dynamics information obtained can provide strong support for the dynamic monitoring and management of forestry carbon sinks, and can help formulate scientific forestry management and carbon emission reduction strategies and optimize resource allocation; in addition, the multidimensional grid model and the polynomial model can be adjusted according to different data and environmental conditions and have good adaptability; the method disclosed in this embodiment can effectively improve the comprehensiveness, accuracy and timeliness of the evaluation.

[0092] Reference Figure 2 , the construction method of the multidimensional grid model specifically includes:

[0093] S211, determining a grid dimension of a target forestry carbon sink project based on the first data set;

[0094] Grid dimensions include time dimension, space dimension, ecological dimension, economic dimension and social dimension.

[0095] Specifically, the first data set is analyzed to determine the main factors that need to be considered, and the appropriate grid dimension is determined based on the analysis results to ensure that all factors that have an impact on the carbon sequestration project are included.

[0096] For example, the time dimension may include project years (such as 150 years); the spatial dimension may include longitude (X-axis), latitude (Y-axis), and altitude (Z-axis); the ecological dimension may include tree species type and forest age; the economic dimension may include carbon price; and the social dimension may include a community participation index.

[0097] It should be noted that the grid dimensions can be flexibly configured according to the needs of the project.

[0098] S212, determining the grid resolution of each type of grid dimension;

[0099] Grid resolution includes temporal resolution, spatial resolution, ecological resolution, economic resolution or social resolution. For temporal resolution, the temporal resolution can be determined according to the time period of the project and the frequency of data updates, such as monthly, quarterly or annually. For spatial resolution, you can set longitude and latitude 0.01° (about 1km), altitude 50m, etc. For ecological resolution, the ecological resolution is determined according to the complexity and monitoring capacity of the ecosystem, such as setting 5 main tree species with a forest age interval of 5 years. For economic resolution, the carbon price interval can be set to 5 US dollars / ton. For social resolution, the corresponding resolution is determined according to the scale of economic activities and the scope of social impact. For example, the community participation index interval can be set to 0.1 (i.e. 01 range).

[0100] This step can help control the amount of data, making it more efficient in processing and use, while ensuring that the level of detail of the data meets actual needs and avoiding data redundancy or information loss caused by too fine or too coarse resolution.

[0101] S213, based on the grid dimension and the grid resolution, corresponding data points in the first data set are allocated to specific positions of the grid according to preset rules to generate a multi-dimensional grid model.

[0102] Specifically, corresponding data points in the first data set are mapped to the nearest grid nodes in order to simplify the complexity of the data into simple coordinates on the grid, which can make prediction and analysis more intuitive and easier to handle.

[0103] Specifically, first determine the location of each data point, for example, determine the latitude and longitude and altitude of the data point so as to map it to the spatial dimension; or simultaneously determine the time (year), ecological information (such as tree species type, forest age), economic information (carbon price) and social information (community participation index) of the data point. Secondly, perform grid node matching, that is, determine the grid node closest to the data point location. Since the grid has a fixed resolution (for example, longitude and latitude 0.01°, altitude 50m), this means that not all data points can fall precisely on the grid node. Therefore, an algorithm can be used to find the grid node closest to the data point. Finally, when the nearest grid node is found, the information of the data point (such as data value, attributes, etc.) is assigned to the node.

[0104] For example, if you have a data point about the charcoal density of trees in a certain area, the location of the data point may not fall exactly on the 1km x 1km x 50m grid you defined. In this case, the data point will be assigned to the spatial grid node closest to its location; if the point is between two adjacent grid nodes, you may need to weight these nodes according to the distance.

[0105] Through this step, complex field data can be simplified into points on the grid, making it easier to perform various operations and analyses on the grid.

[0106] The method for constructing a multidimensional grid model disclosed in this embodiment provides a comprehensive framework for evaluating and managing forestry carbon sink projects to ensure the sustainability and economic benefits of the projects. Through the grid model, the impact of the project can be more accurately predicted and evaluated, providing a scientific basis for decision-making. At the same time, it helps to improve project management efficiency, optimize resource allocation and management strategies through the grid model, and improve the success rate of the project.

[0107] Reference Figure 3 , the method for obtaining the carbon storage polynomial fitting function specifically includes:

[0108] S310, determining a target polynomial basis function based on a multidimensional grid model.

[0109] Specifically include:

[0110] S311, determine the order, that is, the order of the polynomial can be selected according to the complexity of the data, such as 3rd or 4th order. The order of the polynomial determines the complexity of the function that can be represented by the model. A higher order polynomial can fit more complex data changes.

[0111] S312, determining the required data points and their coordinate information based on the multi-dimensional grid model.

[0112] Specifically, the required sample data points are determined in the multidimensional grid model; the multidimensional grid model refers to a grid established in multiple dimensions (such as two-dimensional or three-dimensional space) to represent the distribution of data or functions. These grid points need to be selected according to the requirements of the model and data, and their coordinate information must be recorded.

[0113] S313, determining a target polynomial basis function based on the data points and their coordinate information.

[0114] Specifically, appropriate polynomial basis functions can be selected according to previously obtained data points and their coordinates. Basis functions are basic functions used to construct target polynomials. Common polynomial basis functions include Lagrange basis functions, Chebyshev basis functions, etc. These basis functions will be combined with data points to form the final fitting polynomial.

[0115] For example, for two-dimensional spatial data, you can choose a two-dimensional polynomial, such as:

[0116] P(x,y)=a0+a1x+a2y+a3x 2 +a4xy+a5y 2 .

[0117] In two-dimensional space, the set of cubic polynomial basis functions may include 1, x, y, x 2 ,xy,y 2 ,x 3 ,x 2 y,xy 2 ,y 3 Etc.

[0118] For a 4th order polynomial, the basis function set can be defined as: Where t represents time, x, y, and z represent spatial coordinates, s represents tree species, a represents forest age, p represents carbon price, and c represents community participation.

[0119] S320, obtaining polynomial coefficients based on a least squares method.

[0120] Specifically, a least squares problem is constructed. The goal of the least squares method is to find the optimal polynomial coefficients so that the error between the polynomial model and the actual data is minimized. For example, a mathematical equation or a matrix equation is established. The system of equations obtained by the least squares method is usually linear. Then, by solving the linear equation, the polynomial coefficients can be obtained.

[0121] Furthermore, the linear equations can be solved using the pseudo-inverse or normal equations of the matrix, or numerical methods such as Gaussian elimination, LU decomposition, QR decomposition, etc. can be used to solve these systems of equations, or numerical computing tools (such as Python's NumPy library, MATLAB, etc.) can be used to conveniently solve these systems of equations. The tools will automatically process the inverse of the matrix or other numerical solutions to obtain the polynomial coefficients.

[0122] Specifically, the goal of the least squares method is to find the optimal polynomial coefficients so that the error between the polynomial model and the actual data is minimized.

[0123] Suppose we have a polynomial model such as f(x,y)=a0+a1x+a2y+a3x 2 +a4xy+a5y 2 +……, where a i are the coefficients that need to be fitted.

[0124] Error definition: For each data point (x i ,y i) and the corresponding observation z i , calculate the difference between the model prediction value and the actual observation value, that is, the error ∈ i =z i -f(x i ,y i ).

[0125] Minimization objective: The goal of the least squares method is to minimize the sum of squares of all errors, that is, min∑(∈ i ) 2 , that is, min∑(z i -f(x i ,y i )) 2 .

[0126] S330, obtaining a carbon storage polynomial fitting function based on the polynomial coefficients and the target polynomial basis function.

[0127] For example, f(t,x,y,z,s,a,p,c)=β0+β1t+β2x+β3y+β4z+β5s+β6a+β7p+β8c+β9t 2 +β 10 tx+...+β 287 c 4 .

[0128] In this embodiment, the error between the obtained carbon storage polynomial fitting function and the target polynomial basis function at key points does not exceed 5%.

[0129] The method for obtaining the carbon reserve polynomial fitting function disclosed in this embodiment can more accurately predict carbon reserve changes and improve the prediction ability of the model through multidimensional grid model and polynomial fitting; in this embodiment, the multidimensional grid model combined with polynomial fitting can comprehensively consider data in multiple dimensions such as time, space, ecology, economy and society, provide a comprehensive perspective for carbon reserve analysis, and provide a systematic method to quantify changes in carbon reserves, providing a scientific basis for policy formulation, resource management and project planning. In this embodiment, the polynomial basis function and coefficient calculation can be adjusted according to different data dimensions and complexity, with strong flexibility and adaptability, and can adapt to various practical application scenarios.

[0130] Specifically, this embodiment is described in detail by taking a forestry carbon sink project covering an area of ​​100 square kilometers and having a project period of 30 years as an example.

[0131] 1) Define grid dimensions. The time dimension includes the project year (130 years); the spatial dimension includes longitude (X: 99.5°E to 100.5°E), latitude (Y: 24.5°N to 25.5°N), and altitude (Z: 1000m to 3000m); the ecological dimension includes tree species (Yunnan pine, Armand pine, Yunnan hemlock), forest age (060 years); the economic dimension includes carbon price (US$550 / ton); and the social dimension includes community participation index (01).

[0132] 2) Determine the grid resolution. The temporal resolution is 1 year; the spatial resolution includes 0.01° (about 1 km) of latitude and longitude and 50 m of altitude; the ecological resolution includes 3 main tree species with a forest age interval of 5 years; the economic resolution includes a carbon price interval of 5 US dollars / ton; and the social resolution includes a community participation index interval of 0.1.

[0133] 3) Map data points to the nearest grid nodes

[0134] 4) Considering the complexity of the data, a 4th-order polynomial is selected.

[0135] Define the basis function set: Among them, t represents time, x, y, and z represent spatial coordinates, s represents tree species, a represents forest age, p represents carbon price, and c represents community participation.

[0136] 5) Fitting polynomial coefficients.

[0137] Construct a least squares problem: min||Aβb||2, where A is the basis function matrix, β is the coefficient vector, and b is the observed data vector; use QR decomposition or singular value decomposition (SVD) to solve the linear equations and obtain the polynomial coefficients β.

[0138] 6) Use the obtained polynomial to interpolate in the entire 7-dimensional space.

[0139] Through this process, we get a multivariate polynomial function of the following form:

[0140] f(t,x,y,z,s,a,p,c)=β0+β1t+β2x+β3y+β4z+β5s+β6a+β7p+β8c+β9t 2 +β 10 tx+...+β 287 c 4 , this function can be used to estimate carbon storage at any location and at any point in time within the project area, taking into account the effects of tree species, forest age, carbon price and community participation.

[0141] For example, to estimate the carbon storage of a 20-year-old Yunnan pine tree at 100.2°E longitude, 25.1°N latitude, 2000m above sea level in year 15 of the project, when the carbon price is $30 / ton and the community participation is 0.8, we simply plug these values ​​into the polynomial function to get the result.

[0142] The method disclosed in this embodiment provides a continuous representation of data, allowing us to perform interpolation at unsampled points, compressing the original data and saving storage space, capturing the complex interactions between variables, and providing a smooth function expression for subsequent numerical simulation and optimization.

[0143] Reference Figure 4 , the method for obtaining the polynomial coefficients includes:

[0144] S321, constructing a polynomial matching model based on the target polynomial basis function, where the polynomial matching model includes a plurality of polynomial functions.

[0145] Specifically, we can select appropriate polynomial basis functions according to the dimension and complexity of the data, and select appropriate polynomial forms according to the different characteristics of the data to improve the accuracy of fitting. By constructing multiple polynomial functions, we can better adapt to the complexity and changes of the data.

[0146] S322, determining a least squares target based on the polynomial matching model.

[0147] Specifically, define the least squares objective: set the objective function to be the sum of squared errors of the polynomial function. For example, for a polynomial model, the least squares objective function can be defined as:

[0148] Among them, y i is the actual observed value, P(x i ,z i ,…) are the predicted values ​​of the polynomial function.

[0149] Objective determination: Based on the above definition, the least squares objective is to minimize the sum of squared errors between the actual observations and the polynomial predictions.

[0150] In this step, the definition of the least squares objective function helps to find the polynomial coefficients that minimize the error between the model prediction value and the actual data; by optimizing the sum of squared errors, more accurate polynomial coefficients can be obtained, thereby improving the model fitting effect.

[0151] S323, converting the polynomial matching model into a thread equation group.

[0152] Specifically, each polynomial function in the polynomial matching model is applied to the actual data to form a set of linear equations; these equations are converted into matrix form for solving.

[0153] For example, assuming there are n data points and k basis functions, we can express the polynomial basis function value and the corresponding coefficients of each data point in matrix form.

[0154] Matrix representation includes: using matrix A to represent the values ​​of all basis functions, vector a to represent polynomial coefficients, and vector b to represent observations. Then the linear equations can be represented as Aa=b, where A is an n×k matrix and each row corresponds to the basis function value of a data point. Then converting the problem into a linear equation helps to solve it using linear algebra tools. The equations in matrix form can be quickly solved by computer programs to improve efficiency.

[0155] S324, solving the thread equation group based on the least squares objective to obtain polynomial coefficients.

[0156] Specifically, the linear equations are solved using a least squares method, for example, using a pseudo-inverse of a matrix or other numerical solution method to minimize the sum of squared errors.

[0157] The normal equation is a=(A T A) -1 A T b, where A T is the transpose of the matrix A, (A T A) -1 is the matrix A T The inverse matrix of A, the calculated coefficient vector a is the coefficient of the polynomial function.

[0158] The least squares method provides an efficient method to solve a system of linear equations and obtain the optimal polynomial coefficients. By optimizing the objective function, it can be ensured that the obtained polynomial coefficients can best fit the actual data under given data.

[0159] Through the systematic polynomial matching model and least squares method, the data can be accurately fitted and the predictive ability of the model can be improved. The diversity of the polynomial matching model allows adaptation to different types of data, ensuring the comprehensiveness of the fitting. The application of linear equations and least squares method makes the calculation process of polynomial coefficients efficient and scalable. The obtained polynomial coefficients can provide a scientific basis for further analysis, decision-making and optimization, and support data-driven decision-making processes.

[0160] Reference Figure 5 The method for obtaining carbon dynamic information of the target forestry carbon sink project based on the polynomial model specifically includes:

[0161] S510, constructing a partial differential equation model for the project based on a polynomial model;

[0162] The partial differential equation model of the project includes the dynamic partial differential equations of carbon storage, the dynamic partial differential equations of biomass and the dynamic partial differential equations of soil organic matter.

[0163] Specifically, the carbon storage dynamic partial differential equations include: It describes the rate of change of carbon storage C over time, where B is biomass, S is soil organic matter, t is time, (x, y, z) is spatial coordinate, and s is tree species.

[0164] The set of partial differential equations for biomass dynamics includes: Describes the rate of change of biomass B over time, where C is the carbon stock.

[0165] The soil organic matter dynamic partial differential equations include: Describes the rate of change of soil organic matter S over time.

[0166] Furthermore, if the carbon storage C(x,y,t) depends on the spatial position x, y and time t, a partial differential equation (PDE) is constructed to represent its dynamic changes. For example:

[0167] Among them, D c is the diffusion coefficient, R c are sources of carbon storage (such as plant growth and degradation).

[0168] Correspondingly, the partial differential equation describing the change of biomass B(x,y,t) is:

[0169] Among them, D B is the diffusion coefficient of biomass, R B are the source terms of biomass (such as photosynthesis and respiration).

[0170] Correspondingly, the partial differential equation describing the change of soil organic matter S(x,y,t) is:

[0171] Among them, D S is the diffusion coefficient of soil organic matter, R S It is the source term of soil organic matter (such as decomposition and input).

[0172] S520, approximating partial derivatives of the partial differential equation model of the project based on a finite difference strategy at grid points of the multidimensional grid model to obtain a rate of change of carbon storage.

[0173] Specifically, in the multidimensional grid model, the spatial area is discretized into a number of grid points and the time step is set; an appropriate finite difference strategy (such as forward difference, backward difference or central difference) is selected to approximate the derivatives in the partial differential equation; and then the finite difference strategy is used to calculate the rate of change of carbon storage at the grid points.

[0174] For example, for partial derivatives in one-dimensional space and time, forward differentiation can be used:

[0175]

[0176] The corresponding rate of change of carbon storage is:

[0177]

[0178] In this embodiment, by establishing and solving the partial differential equation model, the dynamic changes of carbon storage, biomass and soil organic matter can be analyzed in detail, providing more in-depth project dynamic information; accurate carbon dynamic information can help formulate scientific forestry carbon sink management and policy strategies, and improve the effectiveness and sustainability of carbon sink projects; the finite difference strategy is applicable to various complex actual situations, including different spatial resolutions and time steps, and has good adaptability. The overall solution provides a systematic framework to evaluate and predict changes in carbon storage, providing a scientific basis for long-term resource management and environmental protection.

[0179] Reference Figure 6 The method for obtaining the rate of change of carbon storage specifically includes the following steps:

[0180] A100, uses a finite difference strategy on the grid points of a multidimensional grid model to approximate the time and space derivatives.

[0181] The time derivative represents the change of model variables (such as carbon storage C) over time. For the time derivative, we usually use the finite difference method for discretization. Consider a discretization process with a time step of Δt: The basic formula is: Among them, C (n) is the variable value at the current time step, C (n+1) is the value of the variable at the next time step.

[0182] Spatial derivatives include:

[0183]

[0184] Among them, h x 、h y 、h z are the grid spacing in the x, y, and z directions respectively.

[0185] A200, obtains the target partial differential equation system based on the project partial differential equation model.

[0186] For example, the target partial differential equation system is Among them, C is carbon storage, NPP is net primary productivity, R is respiratory loss, H is harvesting loss, and D is diffusion coefficient.

[0187] A300, determine the target basis function.

[0188] Specifically, the objective basis function is φ i (x),

[0189] Where h is the density of the grid cells, x i is the position of the grid node.

[0190] A400, constructs discrete equations based on time derivatives, space derivatives, target partial differential equations, target basis functions, and Galerkin method:

[0191] The Neumann boundary condition is: the normal derivative on the boundary is 0.

[0192] Specifically, the discrete equation can be:

[0193] The Neumann boundary condition in this step means that there is no inflow or outflow of carbon at the boundary.

[0194] A500 calculates the carbon storage at the next time step at each grid point based on the discrete equation and obtains the total carbon storage at each time step.

[0195] Among them, the carbon storage in the next time step is C (n+1) ; The total carbon storage at each time step n is Where N x 、N y 、N z are the number of grids in the x, y, and z directions respectively. Furthermore, the carbon storage of all grid points (i, j, k) can be Stored in a multidimensional array or matrix.

[0196] A600 processes the total carbon storage at each time step based on the time series to obtain the dynamic change information of the total carbon storage over time.

[0197] Specifically, time series data are used to draw charts to obtain information on the dynamic changes in total carbon storage over time, including the differences in carbon storage between different tree species and different locations, to help understand the impact of different factors (such as tree species, location, and management measures) on carbon storage.

[0198] Furthermore, the dynamic change information may include growth rate,

[0199] For example, taking the changes in carbon storage of the Yunnan forestry carbon sink project as an example, the specific changes include:

[0200] 1) Define state variables. For example, carbon storage C(t,x,y,z,s): represents the carbon storage of tree species s at time t, location (x,y,z).

[0201] 2) Construct partial differential equations.

[0202] The carbon stock equation includes: Where NPP is net primary productivity, R is respiratory loss, which is a function of carbon storage, H is harvesting loss, which is determined by the management plan, and D is the diffusion term, which represents the movement of carbon in space (such as through roots or litter).

[0203] 3) Define the time step: Δt = 1 year; space step: Δx = Δy ≈ 1 km, Δz = 50 m;

[0204] 4) Approximate partial derivatives at grid points.

[0205] The time derivative approximation includes: in, It represents the value of the variable C at the grid point (i, j, k, s) at time step n, and Δt is the time step, that is, the time interval between two consecutive time steps.

[0206] The spatial derivative approximation includes:

[0207] This formula uses the central difference method to approximate the second-order derivative in space It represents the degree of curvature of variable C in the x-direction of space, that is, the rate of change of variable C with respect to spatial position.

[0208] 5) Use piecewise linear function as a basis function.

[0209] Applying the Galerkin method, we get the discrete equation:

[0210] in, Used to represent changes in the solution at the mesh nodes.

[0211] integral It is performed on the entire computational domain; specifically, the integration in the Galerkin method is used to transform the continuous form of the partial differential equation into a discrete form, and the integral terms include time derivatives, source terms, diffusion terms, etc.

[0212] in, is the time derivative, which is a discretization term in time, indicating the change of the solution from n to n+1 within the time step Δt. This term measures the change of the variable C over time.

[0213] NPP is net primary productivity, which usually represents carbon accumulation or biomass production in an ecosystem, R is metabolic respiration, which represents carbon release from the system, and H is harvest or human intervention, which represents the amount of carbon or biomass removed from the system.

[0214] is the diffusion or diffusion term, which describes the diffusion process of variable C in space. DD is the diffusion coefficient, are the second-order spatial derivatives in the x, y, and z directions respectively.

[0215] This equation is in each basis function The above are all true, which ensures that the discretized partial differential equations satisfy the equations under these basis functions. This is the core idea of ​​the Galerkin method. By making the weighted integral of the residual under each basis function zero, a discrete set of algebraic equations is obtained.

[0216] The Neumann boundary conditions are: Indicates that there is no flow of carbon out of or into the project area.

[0217] 6) Solving, specifically including obtaining the initial carbon storage values ​​set for all grid points based on the carbon storage polynomial fitting function, that is, setting the initial carbon storage values ​​for all grid points at time t=0, and using them as the initial values ​​for all grid points during initialization; for each time step, constructing a linear equation system AX=b, where X is the carbon storage at the next time step, and using an iterative method (such as a conjugate gradient method) to solve the linear equation system.

[0218] By updating the carbon storage value according to the solution results, the change of total carbon storage over time can be calculated, which can be used to analyze the differences in carbon storage of different tree species and different locations.

[0219] Through the scheme disclosed in the above embodiments, the carbon dynamics in forestry carbon sink projects can be accurately simulated, taking into account multiple factors such as spatial heterogeneity, temporal changes and tree species differences, and providing reliable numerical simulation results for evaluating the long-term sustainability of the project. These results can be used to predict future carbon reserves, optimize management strategies, and evaluate the impact of the project on climate change.

[0220] Reference Figure 7 , the method for calculating the carbon storage of the next time step at each grid point based on the discrete equation specifically includes:

[0221] B100, constructing a spatiotemporal multigrid hierarchy.

[0222] Specifically, the spatiotemporal multi-grid hierarchy structure includes time scale hierarchy information and space scale hierarchy information.

[0223] The time scale level information includes: the finest level scale is the monthly scale, the middle level scale is the annual scale, and the coarsest level scale is the 5-year scale.

[0224] The spatial scale level information includes the spatial resolution of the finest layer, the spatial resolution of the middle layer, and the spatial resolution of the coarsest layer.

[0225] For example, set the spatial resolution of the finest layer to a 100m x 100m x 10m grid, the spatial resolution of the middle layer to a 1km x 1km x 50m grid, and the spatial resolution of the coarsest layer to a 10km x10km x 200m grid.

[0226] B200, in the recursive order from fine grid to coarse grid, calculates the residual of the current solution of the discrete equation on the finest grid and converts the residual into the residual on the middle grid; then converts the residual on the middle grid into the residual on the coarsest grid.

[0227] The residual on the fine grid is transformed into the residual on the coarse grid. This step is usually done by restricting the operator to map the high-resolution information to the low-resolution grid.

[0228] Specifically, the residual of the current solution of the discrete equation is calculated on the finest grid, that is, the fine grid residual is: h =f h -A h u h .

[0229] Residual r h Represents the difference between the solution on the fine grid and the actual solution. The smaller the residual, the better the current approximate solution u h The closer to the true solution; f h A is the right-hand side term, defined on a fine grid, and usually represents the source term or the forcing term in the equation; h is a discrete operator, a linear operator obtained on a fine grid (for example, a discretized partial differential equation); u h is the solution vector of the current approximate solution on the fine grid.

[0230] This formula calculates the current approximate solution u on the finest grid. h For the residual r of the actual equation h , if r h Non-zero means the current solution is not accurate enough.

[0231] Convert the residual to the residual on the intermediate grid, that is, restrict the residual to the coarse grid, including: r H represents the residual on the coarse grid (i.e. the middle grid), To constrain the operator, the residual r on the fine grid is h When switching to a coarse grid, commonly used restriction operators include the Full Weighted Average method. We transfer the residual information on the fine grid to the coarse grid. This process is to effectively process the residual on the coarse grid and perform further calculations and optimizations.

[0232] In this embodiment, the full weighted average is used as the restriction operator. Fully weighted average is a common restriction operator used to map fine grid data to coarse grid. The main idea is to estimate the data on the coarse grid by taking a weighted average of the values ​​on the fine grid.

[0233] Specifically, for each point on the coarse grid, a weighted average of the fine grid points it covers is calculated; the weight is typically based on the proportion of fine grid cells that are covered by the coarse grid cells.

[0234] For example, if a coarse grid cell is composed of multiple fine grid cells, the residual of the coarse grid cell can be calculated as follows: Among them, ω i is the weight of each fine grid cell, is the residual corresponding to the fine grid cell, and Area is the area of ​​the coarse grid cell.

[0235] By limiting the residual of the fine grid to the coarse grid, the amount of calculation on the coarse grid can be reduced while maintaining sufficient accuracy. Calculation on a coarse grid is more efficient than calculation on a fine grid, which can speed up the solution process. The multi-grid method can converge to a high-precision solution more quickly by processing errors on grids of different resolutions.

[0236] B300, solved on the coarsest grid.

[0237] Specifically, direct solution methods are usually used to solve linear systems on the coarsest grid, because solving the problem on a coarse grid is simpler and has a lower computational cost.

[0238] Solve on the coarsest grid, the formula includes: A H e H =r H , where A H is the discrete operator (matrix) on the coarse grid, which represents the discretized equation on the coarse grid. It is the discrete operator A obtained from the fine grid. hThe result of transferring the operator to the coarse grid is obtained by limiting the operator; e H is the error vector or update amount on the coarse grid, indicating the error of the current solution on the coarse grid. This vector is the target to be solved and indicates the correction amount of the solution on the coarse grid. H is the residual vector on the coarse grid, which represents the actual residual on the coarse grid, usually by substituting the fine grid residual r h Obtained by restricting to a coarse grid.

[0239] Specifically, construct a coarse grid problem: Construct a linear system of equations A on the coarse grid H e H =r H , which is a smaller system of equations and requires less computation than the system on a fine grid. A direct solver is used to solve this linear system of equations. A direct solver (such as LU decomposition) is an exact method that decomposes the matrix A H Solve the system of equations in a simpler matrix form (such as LU decomposition). Although this method has a higher computational complexity, it can provide an exact solution.

[0240] Larger systems of linear equations can be solved using efficient algorithms and numerical libraries.

[0241] B400, interpolation and correction from coarse grid to fine grid.

[0242] Specifically, the coarse grid solution is interpolated to the fine grid, that is, the solution on the coarse grid is interpolated back to the fine grid. This step converts the solution on the coarse grid into the solution on the fine grid through the interpolation operator.

[0243] Then several relaxation iterations are performed on the fine grid, that is, the solution is relaxed iteratively on the fine grid to further improve the accuracy of the solution.

[0244] The specific formulas include: e H is the error vector on the coarse grid, is the solution correction obtained on the coarsest grid, e h is the error vector on the fine grid, and is the correction amount obtained after interpolation of the coarse grid error. is the interpolation operator, and the error e on the coarse grid is H Interpolation onto a fine grid,The interpolation operator is used to estimate the errors on the coarse grid on the fine grid, so that these error corrections can be applied on the fine grid.

[0245] Update the solution on the fine grid, the formula includes: By taking the current solution u on the fine grid h Add the interpolation error e h, and obtain the updated fine grid solution. This step is to improve the accuracy of the solution by correcting the solution on the fine grid. Finally, several Gauss-Seidel relaxation iterations are performed to eliminate high-frequency errors; Gauss-Seidel iteration is a classic iterative method for solving linear equations, which gradually approaches the exact solution by gradually updating the solution. Relaxation iteration (or iterative correction) is used in this context to further reduce the error on the fine grid; in particular, through Gauss-Seidel iteration, high-frequency errors in the solution (i.e., errors in the details) can be effectively processed and reduced.

[0246] Specifically, the method includes: updating the solution using the error correction amount obtained by interpolation on the fine grid; iterating the updated solution several times using Gauss-Seidel iteration to further reduce the error on the fine grid. These iterative steps gradually eliminate high-frequency errors and improve the accuracy of the solution by updating the solution at each grid point.

[0247] B500, adaptive mesh refinement.

[0248] The purpose of adaptive mesh refinement is to increase the mesh resolution in areas where higher accuracy is required and reduce it in other areas, thereby improving computational efficiency and accuracy.

[0249] Specifically, the error estimation criterion is defined, that is, the local error index is calculated according to the gradient of the solution to evaluate the error size in different regions. Then the grid resolution is dynamically adjusted according to the local error, that is, the grid is refined in the area with large error to improve the accuracy, and the grid is coarsened in the area with small error to reduce the computational cost.

[0250] Specifically, the error estimation criterion formula includes: Among them, η K is the local error index, which represents the error measure on unit K, h K is the size of unit K, usually the side length or diameter of the unit, which represents the geometric size of the unit; is the gradient of carbon storage, indicating the rate of change of carbon storage in unit K; is the norm of the carbon storage gradient on unit K, which quantifies the size of the carbon storage gradient.

[0251] Local error index η K is calculated by the cell size h K and the norm of the carbon stock gradient over the cell This index is used to measure the error metric on unit K. If the local error index is large, it means that the solution of the unit may not be accurate enough and the grid needs to be further refined.

[0252] Dynamically adjust the grid resolution according to the local error, including: if ηK >tolerance, indicating that the error in this area is large. In order to improve the accuracy, it is necessary to refine the grid in and around this unit to better capture the changes in carbon storage. K <0.1×tolerance, that is, when the local error index η K When it is less than 10% of the given tolerance, it means that the error in this area is very small. At this time, the mesh in these areas can be coarsened and the amount of calculation can be reduced by merging adjacent units without significantly affecting the accuracy of the solution.

[0253] Finally, carbon stock changes at different spatial and temporal scales can be calculated and displayed, generating dynamic visualizations of carbon stock evolution.

[0254] By applying the spatiotemporal multigrid method, we are able to efficiently simulate and analyze the carbon dynamics of forestry carbon sink projects at different spatiotemporal scales. This method not only improves computational efficiency, but also provides multi-scale insights, helping us to more comprehensively assess the long-term sustainability of the project. For example, we can simultaneously analyze monthly local carbon uptake fluctuations and annual regional carbon storage trends, providing a scientific basis for project management and policy making.

[0255] This comprehensive numerical simulation framework provides a powerful tool for the sustainability assessment of forestry carbon sink projects, capable of handling complex ecosystem dynamics and providing a reliable data foundation for subsequent artificial intelligence analysis and comprehensive evaluation.

[0256] Reference Figure 8 The carbon dynamic acquisition method of the forestry carbon sink project disclosed in this application also includes:

[0257] C100, obtain historical climate characteristic data, historical economic characteristic data, historical social characteristic data and target ecological characteristic data of the target forestry carbon sink project;

[0258] The target ecological characteristic data include: current carbon storage and annual carbon absorption based on polynomial models, and normalized vegetation index extracted from remote sensing data. The normalized difference vegetation index (NDVI) is a commonly used remote sensing index used to estimate the health and density of vegetation. Specifically, NDVI is calculated from the reflectance of the red and near-infrared bands in remote sensing images, and the formula is: Among them, NIR is the reflectivity of the near-infrared band, and RED is the reflectivity of the red light band.

[0259] Ecological significance: NDVI values ​​range from -1 to 1. Generally, high NDVI values ​​(close to 1) indicate higher vegetation coverage and health, while low NDVI values ​​(close to 0 or negative) indicate sparse vegetation or drought conditions. NDVI can be used to monitor vegetation growth, drought severity, forest cover changes, etc. Indicators extracted from remote sensing data reflect the health and coverage of vegetation, which helps to understand the growth status and changing trends of ecosystems.

[0260] Historical climate characteristic data may include the annual average temperature and precipitation from meteorological stations, and the frequency of extreme climate events such as drought and floods; historical economic characteristic data may include carbon prices (from carbon trading market data), project operating costs (from project financial reports), etc.; historical social characteristic data may include local community participation (from social surveys), changes in employment rates (from government statistics), etc.

[0261] C200,selects several key features from each type of feature data to obtain the key data set of the corresponding type.

[0262] C300, determine a number of initial learning models, the number of which is consistent with the number of types of feature data.

[0263] Specifically, the initial learning model includes any one of a random forest model, a gradient boosting tree model, and a long short-term memory network model.

[0264] C400, based on the initial learning model corresponding to each type of key data set training, adopts the cross-validation strategy to verify the trained initial learning model, and uses the Bayesian optimization strategy to tune the parameters of the initial learning model that meets the verification requirements. The tuned initial learning model is the target learning model.

[0265] Specifically, the corresponding key data set is divided into a training set (70%), a validation set (15%) and a test set (15%), where the training set is used for model training, the validation set is used to adjust the model hyperparameters, and the test set is used to evaluate the final performance of the model, that is, the model is trained by using the training set data, and the model parameters are adjusted on the validation set to improve its performance.

[0266] In this step, a cross-validation strategy is used to verify the initial learning model after training, including: dividing the dataset into k folds, k-1 folds are used to train the model, and the remaining fold is used to test the model. This process is repeated k times, and each fold is used as a test set. The final model performance is the average of all k test results.

[0267] In this step, the Bayesian optimization strategy is used for parameter tuning. Specifically, by establishing a probabilistic model between hyperparameters and model performance, the hyperparameter combination that is most likely to improve model performance is intelligently selected. This method is usually more efficient than grid search or random search because it uses previous test results to guide the search process.

[0268] C500 obtains prediction results of corresponding types based on the first data set and several target learning models.

[0269] C600 uses a weighted average integration strategy to process several prediction results to obtain target results, which include long-term trend prediction information of carbon storage, project economic feasibility prediction information and ecosystem service value prediction information.

[0270] For example, long-term sustainability forecasts that can be generated include: predicting annual carbon uptake and cumulative carbon storage over the next 30 years, assessing the economic feasibility of projects under different carbon price scenarios, and predicting long-term trends in the value of ecosystem services.

[0271] The method disclosed in this embodiment can comprehensively consider the factors affecting carbon storage and carbon absorption by integrating historical climate, economic, social and ecological data. This comprehensiveness helps to more accurately model and predict carbon dynamics; using polynomial models to obtain carbon storage and annual carbon absorption, as well as the normalized vegetation index extracted from remote sensing data, can more finely capture ecosystem changes and carbon storage dynamics; screening key features from each type of feature data helps to reduce the complexity of the model and focus on the features that have the greatest impact on the prediction results, thereby improving the accuracy and stability of the model; by training multiple initial learning models and using cross-validation and Bayesian optimization strategies to tune their parameters, the most suitable data model can be found, thereby improving the reliability and accuracy of the prediction results; using a weighted average integration strategy to synthesize the prediction results of multiple models into a comprehensive prediction result can reduce possible deviations in a single model and improve the comprehensiveness and accuracy of the prediction; the target results obtained include information on long-term trends in carbon storage, project economic feasibility, and ecosystem service value, which helps to comprehensively evaluate the effectiveness and feasibility of forestry carbon sink projects.

[0272] This method can dynamically adjust models and forecasting strategies to adapt to new data and changes. This adaptability enables the project to obtain effective forecasting and analysis results under different environments and conditions; by integrating the forecast results of multiple models and using optimization techniques, the forecast errors of a single model can be effectively corrected, enhancing the fault tolerance and robustness of the system. At the same time, it provides scientific forecasting information on carbon storage, economic feasibility and ecological service value, which helps policymakers, investors and managers make more scientific and reasonable decisions. In addition, the program can evaluate the long-term benefits of the project, including environmental and economic benefits, and help optimize the implementation strategies and management measures of the project.

[0273] This method provides accurate carbon dynamics predictions for forestry carbon sink projects through the integration of multiple data sources, key feature screening, model optimization and prediction integration. It not only improves the accuracy and comprehensiveness of the predictions, but also enhances the adaptability and flexibility of the model, ultimately supporting scientific decision-making and project evaluation. These benefits give this method significant advantages in the management and evaluation of forestry carbon sink projects.

[0274] Reference Fig. 9 The second aspect of the present application discloses a method for evaluating the long-term sustainability of a forestry carbon sink project. The method is based on the carbon dynamic acquisition method of the forestry carbon sink project and further includes:

[0275] S10, based on the carbon dynamic information, obtains the annual carbon absorption and cumulative carbon storage within a future preset period, and creates a time series prediction chart.

[0276] Based on the acquired data, calculate the annual carbon absorption and cumulative carbon storage in the future preset period (such as the next 10 years, 20 years, etc.). In order to better understand the changing trends of these data, we will create a time series forecast chart that shows the changes in carbon absorption and carbon storage in the future years.

[0277] Example: Assume that data from the beginning of the project shows that the average annual carbon absorption of trees is 500 tons. According to the growth model, the annual carbon absorption may increase in the next 10 years. We plot these predicted values ​​into a time series graph to show the trend of future carbon absorption.

[0278] S20, heat map of sensitivity analysis to determine the economic feasibility of the project based on the time series forecast graph.

[0279] Use time series forecast charts to conduct sensitivity analysis of the project's economic feasibility. The purpose of this step is to understand the impact of different factors (such as tree growth rate, market carbon price, forest management costs, etc.) on the project's economic benefits. Create a sensitivity analysis heat map to observe their impact on the project's economic feasibility by changing one or more key parameters; the generated heat map will show the sensitivity of economic benefits to different parameter changes, such as the impact of carbon market price fluctuations on the project's net present value (NPV).

[0280] Example: If changes in carbon market prices can significantly affect the economic benefits of a project, we will use a heat map to show the extent to which price changes affect the project's economic indicators. This helps identify which economic factors are most sensitive to project feasibility.

[0281] S30, identify the key influencing factors in the target forestry carbon sink project, and obtain the project's long-term sustainability assessment information based on the key influencing factors and sensitivity analysis heat map.

[0282] Based on the sensitivity analysis heat map, key factors that have a significant impact on the long-term sustainability of the project are identified, which may include environmental conditions, management measures, policy changes, etc. Combined with the identified key influencing factors, the long-term sustainability of the project is evaluated, which can be achieved by developing adaptation strategies, optimizing management measures, adjusting budgets, etc.

[0283] Example: Assume that a sensitivity analysis shows that climate change has a significant impact on tree growth and carbon storage. To ensure the long-term sustainability of the project, it may be necessary to consider introducing climate-resilient tree species or optimizing forest management strategies.

[0284] Through the above methods, we can comprehensively evaluate the long-term sustainability of forestry carbon sink projects. This includes starting with carbon dynamics data, conducting time series forecasts, further determining the key factors of economic feasibility through sensitivity analysis, and formulating long-term sustainability assessment strategies based on these factors. This method can not only help optimize project implementation, but also provide important decision-making basis for policymakers and investors.

[0285] The present application also discloses a carbon dynamic acquisition system for a forestry carbon sink project, comprising:

[0286] A collection module is used to collect multi-source data of the target forestry carbon sink project and pre-process the multi-source data to obtain a first data set; the multi-source data includes ecological data, climate data, economic data and social data;

[0287] A construction module, configured to construct a multidimensional grid model based on the first data set;

[0288] A fitting function acquisition module, used to obtain a carbon storage polynomial fitting function based on a multidimensional grid model and a preset strategy;

[0289] An interpolation module, used for interpolating within the entire network of the multidimensional grid model using a carbon storage polynomial fitting function to obtain a polynomial model;

[0290] The analysis module is used to obtain the carbon dynamic information of the target forestry carbon sink project based on the polynomial model.

[0291] The methods in the carbon dynamic acquisition method for the forestry carbon sink project disclosed in this application are all applicable to the solutions in the carbon dynamic acquisition system for the forestry carbon sink project disclosed in this embodiment, so they will not be described one by one here.

[0292] The computer device according to the embodiment of the present disclosure includes a memory and a processor. The memory is used to store non-temporary computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0293] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the computer device to perform desired functions. In one embodiment of the present disclosure, the processor is used to run the computer-readable instructions stored in the memory, so that the computer device executes all or part of the steps of the carbon dynamic acquisition method for the forestry carbon sink project or the long-term sustainability assessment method for the forestry carbon sink project of each embodiment of the present disclosure.

[0294] Those skilled in the art should be able to understand that in order to solve the technical problem of how to obtain a good user experience, the present embodiment may also include well-known structures such as a communication bus and an interface, and these well-known structures should also be included in the protection scope of the present disclosure.

[0295] like Fig.10 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure is shown, which is a schematic diagram of the structure of a computer device suitable for implementing the embodiment of the present disclosure. Fig.10 The computer device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0296] like Fig.10As shown, the computer device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the computer device are also stored. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0297] Typically, the following devices can be connected to the I / O interface: input devices such as sensors or visual information acquisition devices; output devices such as display screens; storage devices such as tapes, hard disks, etc.; and communication devices. The communication device can allow the computer device to communicate with other devices (such as edge computing devices) wirelessly or by wire to exchange data. Fig.10 A computer device having various devices is shown, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0298] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processor, all or part of the steps of the carbon dynamic acquisition method of a forestry carbon sink project or the long-term sustainability assessment method of a forestry carbon sink project of an embodiment of the present disclosure are executed.

[0299] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0300] According to the computer-readable storage medium of the embodiment of the present disclosure, non-transitory computer-readable instructions are stored thereon. When the non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the carbon dynamic acquisition method of the forestry carbon sink project or the long-term sustainability assessment method of the forestry carbon sink project of each embodiment of the present disclosure are executed.

[0301] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).

[0302] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0303] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0304] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, devices, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0305] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0306] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0307] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0308] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0309] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for dynamically acquiring carbon in a forestry carbon sink project, characterized in that: include: Collecting multi-source data of a target forestry carbon sink project, and preprocessing the multi-source data to obtain a first data set; The multi-source data include ecological data, climate data, economic data and social data; constructing a multidimensional grid model based on the first data set; Obtaining a carbon storage polynomial fitting function based on the multidimensional grid model and a preset strategy; interpolating the carbon storage polynomial fitting function within the entire network of the multidimensional grid model to obtain a polynomial model; The carbon dynamic information of the target forestry carbon sink project is obtained based on the polynomial model.

2. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 1, characterized in that: The constructing a multidimensional grid model based on the first data set includes: Determining a grid dimension of the target forestry carbon sink project based on the first data set; The grid dimensions include time dimension, space dimension, ecological dimension, economic dimension and social dimension; determining a grid resolution for each type of said grid dimension; The grid resolution includes temporal resolution, spatial resolution, ecological resolution, economic resolution or social resolution; Based on the grid dimension and the grid resolution, corresponding data points in the first data set are allocated to specific positions of the grid according to preset rules to generate the multi-dimensional grid model.

3. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 1, characterized in that: The step of obtaining a carbon storage polynomial fitting function based on the multidimensional grid model and a preset strategy includes: Determining a target polynomial basis function based on the multidimensional grid model; Obtain polynomial coefficients based on the least squares method; A carbon storage polynomial fitting function is obtained based on the polynomial coefficients and the target polynomial basis function.

4. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 3, characterized in that: The obtaining of polynomial coefficients based on the least squares method includes: Constructing a polynomial matching model based on the target polynomial basis function, wherein the polynomial matching model includes a plurality of polynomial functions; determining a least squares objective based on the polynomial matching model; converting the polynomial matching model into a set of thread equations; The thread equation group is solved based on the least squares objective to obtain polynomial coefficients.

5. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 1, characterized in that: The step of obtaining carbon dynamic information of the target forestry carbon sink project based on the polynomial model includes: Constructing a partial differential equation model of the project based on the polynomial model; The partial differential equation model of the project includes a group of dynamic partial differential equations of carbon storage, a group of dynamic partial differential equations of biomass and a group of dynamic partial differential equations of soil organic matter; The partial derivatives of the partial differential equation model of the project are approximated based on the finite difference strategy at the grid points of the multidimensional grid model to obtain the rate of change of carbon storage.

6. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 5, characterized in that: The method of approximating partial derivatives of the partial differential equation model of the project based on a finite difference strategy at the grid points of the multidimensional grid model to obtain the rate of change of carbon reserves includes: using a finite difference strategy on grid points of the multidimensional grid model to approximate time derivatives and space derivatives; Obtaining a target partial differential equation group based on the project partial differential equation model; Determine the target basis function; Based on the time derivative, the space derivative, the target partial differential equation group, the target basis function, and the Galerkin method, a discrete equation is constructed: The boundary conditions are: the normal derivative on the boundary is 0; Calculating the carbon storage at the next time step at each grid point based on the discrete equation, and obtaining the total carbon storage at each time step; The total carbon storage at each time step is processed based on the time series to obtain the dynamic change information of the total carbon storage over time.

7. The method for dynamically acquiring carbon in a forestry carbon sink project according to claim 1, characterized in that: Also includes: Obtain historical climate characteristic data, historical economic characteristic data, historical social characteristic data and target ecological characteristic data of the target forestry carbon sink project; The target ecological characteristic data include: current carbon storage and annual carbon absorption obtained based on the polynomial model, and normalized vegetation index extracted from remote sensing data; A number of key features are selected from each type of feature data to obtain a key data set of the corresponding type; Determine a number of initial learning models, the number of the initial learning models being consistent with the number of types of feature data; Based on the initial learning model corresponding to each type of the key data set training, the trained initial learning model is verified by using a cross-validation strategy, and the parameters of the initial learning model that meets the verification requirements are tuned based on a Bayesian optimization strategy, and the tuned initial learning model is the target learning model; Based on the first data set and the plurality of target learning models, obtaining prediction results of corresponding types; A weighted average integration strategy is used to process a number of the prediction results to obtain target results, which include carbon storage long-term trend prediction information, project economic feasibility prediction information and ecosystem service value prediction information.

8. A method for evaluating the long-term sustainability of a forestry carbon sink project, characterized in that: The method is based on the carbon dynamic acquisition method of the forestry carbon sink project according to any one of claims 1 to 7, and further includes: Based on the carbon dynamic information, the annual carbon absorption and cumulative carbon storage within a future preset period are obtained, and a time series prediction chart is created; Determine a sensitivity analysis heat map of the economic feasibility of the project based on the time series forecast map; Identify the key influencing factors in the target forestry carbon sink project, and obtain the project's long-term sustainability assessment information based on the key influencing factors and the sensitivity analysis heat map.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and, YY+242221P a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the carbon dynamic acquisition method for a forestry carbon sink project described in any one of claims 1 to 7 or the long-term sustainability assessment method for a forestry carbon sink project described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which are used to enable a computer to execute the carbon dynamic acquisition method for a forestry carbon sink project described in any one of claims 1 to 7 or the long-term sustainability assessment method for a forestry carbon sink project described in claim 8.

Images