Life cycle measurement method and system for forestry carbon sink project based on digital twinning

Abstract

The application provides a forestry carbon sink project full life cycle measurement method and system based on digital twinning, relates to forestry carbon sink measurement, and comprises the following steps: acquiring multi-source heterogeneous basic data such as forest three-dimensional skeleton topology, carbon dioxide flux time sequence, mycorrhizal network distribution and meteorological data, coupling forest digital gene and ecological process tensor to generate a digital gene tensor field; then obtaining a double-path carbon flux field of dominant conduit transmission and implicit mycorrhizal transfer through carbon transmission inversion driven by a conduit bundle network; constructing a carbon sink phase change field based on a stability boundary constraint by analyzing a carbon sink phase change critical point; obtaining a carbon re-allocation map regulated by a biological network through a cross-scale carbon re-allocation path deduction; and finally outputting an optimized and adjusted carbon sink evaluation value through full life cycle carbon sink entropy optimization measurement processing. The application realizes the transformation from static estimation to dynamic simulation of carbon sink measurement, and significantly improves the accuracy and reliability of full life cycle evaluation of a carbon sink project.

Description

Technical Field

[0001] This invention relates to the field of forestry carbon sequestration technology, and more specifically, to a method and system for measuring the entire life cycle of forestry carbon sequestration projects based on digital twins. Background Technology

[0002] Currently, forestry carbon sequestration is an important technical means to address climate change. Significant progress has been made in remote sensing monitoring and model simulation in recent years. Existing methods mainly rely on a technical approach that combines remote sensing biomass retrieval with ground sampling surveys to estimate regional carbon sequestration capacity by establishing forest growth equations and carbon storage conversion factors. However, this method, based on static correlation and empirical parameterization, is difficult to effectively capture the dynamic cycling process of carbon in the forest-soil-atmosphere interface. In particular, it cannot quantify the regulatory role of underground mycorrhizal networks-mediated implicit carbon flows on system stability. This leads to significant uncertainties in carbon sequestration results when dealing with extreme climate events and disturbance scenarios, and lacks the ability to warn of abrupt changes in carbon sequestration function. This restricts the accurate accounting and risk management of carbon sequestration projects at the entire life cycle scale.

[0003] Therefore, there is an urgent need for a digital twin-based method and system for measuring the entire life cycle of forestry carbon sink projects to solve the above-mentioned technical problems. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for measuring the entire life cycle of forestry carbon sink projects based on digital twins, in order to improve the aforementioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows:

[0005] Firstly, this application provides a method for measuring the entire life cycle of forestry carbon sink projects based on digital twins, including:

[0006] Acquire multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, distribution map of mycorrhizal network and time series data of meteorology.

[0007] The multi-source heterogeneous basic data is coupled with forest digital genes and ecological process tensors. In this process, features are extracted from the multi-source heterogeneous basic data and the extracted features are mapped to generate a digital gene tensor field.

[0008] The digital gene tensor field is subjected to carbon transport inversion driven by vascular bundle network. Specifically, by decomposing the digital gene tensor field into carbon flux modes that are transported along the vascular bundle network, a dual-path carbon flux field containing dominant vascular transport and recessive mycorrhizal transfer is obtained.

[0009] The dual-path carbon flux field is subjected to carbon sink phase transition critical point analysis processing. Specifically, by identifying the critical conditions for abrupt changes in carbon storage state and quantifying the influence of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated.

[0010] Based on the carbon sink phase transition field, cross-scale carbon redistribution path deduction was performed to obtain a carbon redistribution map based on biological network regulation.

[0011] Based on the carbon redistribution map, the entropy of the carbon sink throughout the entire life cycle is optimized and measured to obtain the optimized and adjusted carbon sink estimate for the entire life cycle.

[0012] Secondly, this application also provides a digital twin-based full life-cycle measurement system for forestry carbon sink projects, including:

[0013] The acquisition unit is used to acquire multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, distribution map of mycorrhizal network and time series data of meteorology.

[0014] The processing unit is used to couple the multi-source heterogeneous basic data with forest digital genes and ecological process tensors. Specifically, by extracting features from the multi-source heterogeneous basic data and mapping the extracted features, a digital gene tensor field is generated.

[0015] The inversion unit is used to perform carbon transport inversion processing driven by vascular bundle network on the digital gene tensor field. In this process, by decomposing the digital gene tensor field into carbon flux modes that are transported along the vascular bundle network, a dual-path carbon flux field containing dominant vascular transport and recessive mycorrhizal transfer is obtained.

[0016] The analysis unit is used to perform carbon sink phase transition critical point analysis processing on the dual-path carbon flux field. Specifically, by identifying the critical conditions for abrupt changes in carbon storage state and quantifying the influence of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated.

[0017] The deduction unit is used to perform cross-scale carbon redistribution path deduction based on the carbon sink phase transition field to obtain a carbon redistribution map based on biological network regulation.

[0018] The estimation unit is used to perform full life cycle carbon sink entropy optimization measurement based on the carbon redistribution map to obtain the optimized and adjusted full life cycle carbon sink estimate.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention constructs a dynamic coupling mechanism between forest digital genes and ecological process tensors, realizing the dual-pathway carbon flux inversion of dominant vascular transport and recessive mycorrhizal transfer. It then establishes a carbon redistribution model based on biological network regulation, ultimately forming a full life-cycle entropy optimization measurement system with carbon sink phase transition early warning capability. This effectively solves the fundamental problem that traditional methods cannot quantify the dynamics of underground carbon flow and its impact on system stability.

[0021] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the process for the whole life cycle measurement method of forestry carbon sink projects based on digital twins as described in this embodiment of the invention;

[0024] Figure 2 This is a schematic diagram of the full life-cycle metering system for forestry carbon sink projects based on digital twins, as described in this embodiment of the invention.

[0025] In the diagram: 701, Acquisition Unit; 702, Processing Unit; 703, Inversion Unit; 704, Analysis Unit; 705, Deduction Unit; 706, Estimation Unit. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0028] Example 1:

[0029] This embodiment provides a method for measuring the entire life cycle of forestry carbon sink projects based on digital twins.

[0030] See Figure 1 The figure shows that the method includes steps S1, S2, S3, S4 and S5.

[0031] Step S1: Obtain multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, mycorrhizal network distribution map and meteorological time series data.

[0032] It is understandable that step S1 constructs a digital twin base by collecting multi-source heterogeneous basic data of the target forest area. The three-dimensional skeleton topology data of the trees is extracted from lidar point clouds. The topology network of branches and trunks is generated by the skeleton extraction algorithm in the existing technology, which is specifically represented by the geometric relationship of vascular bundle channels connecting the root system to the leaf crown. The carbon dioxide flux time series data comes from high-frequency monitoring of the eddy covariance flux tower, which records the net carbon exchange dynamics of the canopy-atmosphere interface. The mycorrhizal network distribution map is obtained by constructing the spatial connection topology of fungal hyphae between the tree roots based on mycelial tracing experiments. The meteorological time series data is collected by the forest microclimate station, including canopy temperature and humidity, precipitation penetration, etc.

[0033] These multi-source data are highly complementary in the forestry carbon sink scenario: the topology of the tree skeleton revealed by lidar point clouds provides a physical channel for subsequent vascular bundle carbon transport modeling, while the mycorrhizal network map breaks through the limitation of traditional carbon measurement neglecting underground biological channels, and provides a unique observation dimension for analyzing hidden carbon flows by quantifying the carbon exchange potential between trees connected by fungal hyphae.

[0034] Step S2: Couple the multi-source heterogeneous basic data with forest digital genes and ecological process tensors. Specifically, feature extraction is performed on the multi-source heterogeneous basic data, and the extracted features are mapped to generate a digital gene tensor field.

[0035] This step involves feature extraction and structured recombination of multi-source heterogeneous basic data: the three-dimensional skeletal topological data of forest trees is transformed into high-dimensional morphological gene vectors describing trunk curvature and branching patterns through topological persistence analysis; and the mycorrhizal network distribution map is encoded into network connectivity feature vectors through graph Laplacian feature mapping. Based on this, tensor product operations are used to fuse the morphological gene vectors, mycorrhizal network feature vectors, and carbon dioxide flux time-series data in a unified multi-dimensional space. This space uses spatial grids, time axes, and ecological process dimensions (such as photosynthesis and respiration consumption) as coordinate bases. This coupling method overcomes the limitations of separating structural and process data in traditional forestry carbon sink models, allowing the morphological characteristics of individual trees (such as canopy complexity affecting photosynthetic efficiency) and the topology of underground mycorrhizal networks (such as hyphal connectivity affecting carbon allocation paths) to directly participate in the calculation of carbon cycle processes.

[0036] Step S3: Perform carbon transport inversion processing driven by vascular bundle network on the digital gene tensor field. Specifically, by decomposing the digital gene tensor field into carbon flux modes that are transported along the vascular bundle network, a dual-path carbon flux field containing dominant vascular transport and recessive mycorrhizal transfer is obtained.

[0037] This approach, understandably, leverages the coupled information of forest structure and ecological processes contained within the digital gene tensor field. First, it reconstructs the continuous transport channels of the forest vascular bundle network using fractal geometry theory, transforming the branch topology encoded in the tensor field into a vessel network model with pore size distribution characteristics. Based on this, mycorrhizal network distribution data is introduced to establish a cooperative transport equation driven by the water potential difference within the vessels and the implicit carbon flow driven by the concentration gradient of the mycelial network. Variational principles are used to perform modal decomposition of the tensor field, reversing the transport flux of photosynthetic products ascending along the xylem vessels (the explicit path) and the carbon exchange flux transferred between trees via ectomycetes (the implicit path). This process overcomes the limitation of traditional carbon econometrics treating trees as isolated individuals, providing a theoretical basis for accurately analyzing the carbon redistribution process in forest ecosystems by quantifying the "wood network" effect formed by underground mycorrhizal networks.

[0038] Step S4: Perform carbon sink phase transition critical point analysis processing on the dual-path carbon flux field. In this process, by identifying the critical conditions for abrupt changes in carbon storage state and quantifying the effect of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated.

[0039] This step, understandably, establishes a carbon storage response model within a non-equilibrium thermodynamic framework by characterizing carbon transport dynamics based on a dual-pathway carbon flux field. The flux ratio of visible vascular transport to recessive mycorrhizal transfer is used as a system order parameter, and the Lyapunov exponential spectrum analysis method is employed to quantitatively describe the stability characteristics of carbon storage under climate disturbances. Specifically, by introducing network percolation theory, the topological connectivity of the mycorrhizal network is transformed into a regulating factor for the system's phase transition threshold. When external stresses (such as drought and pests) lead to a decrease in vascular transport efficiency, the model can quantify the buffering strength of the mycelial network in enhancing inter-tree carbon sharing to improve system stability, thereby accurately analyzing the critical conditions for the carbon sink function to transition from steady state to abrupt change. This treatment, for the first time, achieves a digital characterization of the regulatory capacity of biological networks in forestry carbon sink metering, providing a dynamic stability criterion for the risk assessment of carbon sink projects.

[0040] Step S5: Based on the carbon sink phase transition field, perform cross-scale carbon redistribution path deduction processing to obtain a carbon redistribution map based on biological network regulation;

[0041] This step, as is understandable, uses the stability boundary identified in the carbon sink phase transition field as a constraint condition, and maps the phase transition critical point onto the mycorrhizal network topology to identify vulnerable node sets. Based on random walk theory, it constructs a probability model of carbon particle transport in the hyphal network, and analyzes the evolution of carbon redirection paths under different perturbation scenarios by combining network seepage thresholds. Specifically, it introduces a renormalization group method to maintain the invariance of carbon allocation relationships from microscopic hyphal connections to macroscopic stand-scale carbon distribution, ultimately generating a carbon redistribution map that simultaneously reflects the regulatory capacity of biological networks and the risk of phase transition. This processing, by quantifying the carbon flow recombination effect caused by the failure of vulnerable nodes, breaks through the simplified characterization of network-based carbon transport dynamics in traditional carbon econometrics, providing a spatially explicit analytical tool for predicting the resilience of carbon sink functions under extreme events.

[0042] Step S6: Based on the carbon redistribution map, perform full life cycle carbon sink entropy optimization measurement to obtain the optimized and adjusted full life cycle carbon sink estimate.

[0043] Understandably, the carbon flow path constraints defined in the carbon redistribution map in this step are constructed by modeling the carbon storage distribution at the beginning and end of the project as a probability measure in Wasserstein space, thus building a metric manifold that considers carbon leakage paths. Based on this, the entropy-regularized optimal transport theory is introduced to transform traditional carbon sink measurement into solving an optimal transport problem. By solving, a smooth transport potential function is obtained, thereby achieving a regularized fit to the cross-lifecycle carbon redistribution process while maintaining the physical conservation of carbon flux. This processing effectively suppresses numerical oscillations caused by model nonlinearity through entropy constraints, ensuring that the carbon sink estimate conforms to both the transport constraints of the biological network and the second law of thermodynamics, ultimately generating a physically consistent full-lifecycle carbon sink quantification result.

[0044] Furthermore, step S2 includes steps S21, S22 and S23.

[0045] Step S21: The morphological gene vector extraction process is performed on the three-dimensional skeleton topology data of the forest trees. Specifically, by performing topological persistence analysis on the point cloud skeleton, the trunk curvature data, branch fractal dimension data and canopy filling data are extracted and encoded into high-dimensional feature vectors to obtain the forest tree morphological gene vectors.

[0046] Step S22: Based on the tree morphological gene vector and mycorrhizal network distribution map, perform underground carbon transport channel coupling processing. In this process, the mycorrhizal network map is transformed into a weighted directed graph structure, and the morphological gene vector is projected onto the network spectral domain space using the Laplace eigenmap in graph theory to obtain the tree-mycorrhizal joint gene unit.

[0047] Step S23: Tensor field construction processing is performed based on the forest tree-mycorrhizal joint gene unit and carbon dioxide flux time series data. Specifically, a digital gene tensor field with multi-resolution characteristics is constructed by performing tensor product operation on the singular spectrum of the joint gene unit and the flux time series.

[0048] Understandably, in this step, step S21 calculates the Euclidean distance between point cloud data and generates α-complex filtering to calculate the survival period of topological features (such as connected branches and ring structures) at different scales, thereby extracting topological invariants reflecting the trunk bending pattern. For the complex morphological features of the tree skeleton in forestry scenarios, fractal geometry theory is further used to calculate the fractal dimension of the branching structure, and box counting is used to quantify the self-similarity characteristics of branch distribution. Simultaneously, point cloud density analysis is used to analyze canopy filling characteristics, and the difference coefficient between the convex hull volume and the actual point cloud distribution is calculated. These feature parameters are dimensionality-reduced and fused through principal component analysis. The resulting high-dimensional morphological gene vector not only preserves the geometric features of the trees but also encodes the structural adaptation information formed during their growth. The technical effect of this processing is to transform discrete point cloud observations into feature representations with biophysical significance, providing fundamental data representations that can simultaneously reflect the structural features and physiological functions of trees for subsequent carbon allocation model establishment.

[0049] The formula for calculating the lifespan of topological features is as follows:

[0050]

[0051] Where Pers(f) represents the lifespan of the topological features (connected branches and ring structures) of the point cloud skeleton at different scales, and α death α is the filtering scale for the vanishing topological features of the point cloud skeleton. birthThe filtering scale for the topological features of the point cloud skeleton.

[0052] The formula for calculating the canopy filling characteristics is as follows:

[0053] ;

[0054] Where Cf is the canopy filling difference coefficient, defined as the deviation of the volume ratio from 1. A value close to 0 indicates uniform filling, and it is used in forestry to indicate canopy density. V actual V represents the estimated volume of the actual point cloud distribution, obtained by integrating the point cloud density, and signifies the true filling volume of the leaf cluster. convex The point cloud convex hull volume represents the outer envelope volume of the canopy, reflecting the maximum possible spatial extent of the canopy.

[0055] Next, step S22 first transforms the mycorrhizal network distribution map into a weighted directed graph structure, where nodes represent individual trees, edges represent hyphal connections, and edge weights are determined by hyphal density detection values ​​and spatial distance calculations. An exponential decay model is used to quantify connection strength, and the formula for calculating connection strength is shown below:

[0056] ;

[0057] Among them, W ij ρ represents the strength of mycorrhizal connections. ij Hyphae density for molecular biological detection, d ij λ represents the Euclidean distance between mycorrhizae, and λ is the characteristic attenuation length determined according to the mycorrhizal type.

[0058] Then, based on this, a graph Laplacian operator based on morphological similarity constraints is introduced, the formula of which is shown below:

[0059] ;

[0060] Among them, L m The graph Laplacian operator is based on morphological similarity constraints. D is the degree matrix, M is the morphological similarity matrix, and the similarity between forest tree morphological gene vectors is calculated by Gaussian kernel function. W is the edge weight matrix. The spectral basis function is obtained by spectral decomposition of the Laplacian matrix, and the original morphological gene vectors are projected into the low-dimensional spectral domain space to form joint gene units.

[0061] The formula for projecting the original morphological gene vector into the low-dimensional spectral domain space is shown below:

[0062] ;

[0063] Among them, y i U represents the projected joint gene unit. Tx represents the transpose of the eigenvector matrix. i This is the original morphological gene vector, representing the morphological characteristics of individual trees.

[0064] The unique advantage of step S22 lies in the fact that it achieves nonlinear coupling between aboveground morphological features and underground mycorrhizal networks through morphologically constrained spectral embedding. Compared with traditional graph embedding methods, it can better maintain the biological rationality of carbon transport. It constructs a joint representation feature that can simultaneously encode the morphological differences of individual trees and the topological relationship of mycorrhizal networks, providing a mathematical basis with a clear ecological interpretation for subsequent quantification of implicit carbon transport through hyphal networks.

[0065] Finally, step S23 employs a multi-data fusion method based on Tucker tensor decomposition to couple the forest tree-mycorrhizal joint gene units with the carbon dioxide flux time-series data in a high-order manner. First, singular spectrum analysis is performed on the carbon dioxide flux time-series data, and the main oscillation modes are extracted through singular value decomposition of the trajectory matrix to obtain the time-frequency feature matrix of the flux data. Then, a third-order core tensor is constructed, and the joint gene units are embedded as constraints into the initialization of the gene feature basis matrix using a Tucker decomposition model. The Tucker decomposition model is shown below:

[0066] ;

[0067] Where X is the original data tensor, G is the core tensor, U is the spatial basis matrix (forest distribution), V is the gene feature basis matrix, H is the time basis matrix, X1 represents the modular product operation in the first dimension, X2 represents the modular product operation in the second dimension, and X3 represents the modular product operation in the third dimension.

[0068] This step achieves natural coupling of spatial-genetic-temporal three-dimensional information through tensor decomposition, which better preserves the nonlinear structure of the data compared to traditional matrix decomposition. It constructs a digital gene tensor field that can simultaneously encode individual tree characteristics, mycorrhizal network relationships, and carbon flux dynamics, providing a unified data representation framework for subsequent carbon transport inversion.

[0069] Furthermore, step S3 includes steps S31, S32, and S33.

[0070] Step S31: Reconstruct the xylem vessel network based on the digital gene tensor field. This involves analyzing the vessel bundle connection relationships contained in the three-dimensional skeleton topology of the forest in the tensor field and reconstructing a continuous vessel network model from the root system to the canopy based on fractal geometry theory to obtain a network topology map containing vessel pore size distribution and connectivity characteristics.

[0071] Step S32: Perform dual-path carbon transport coupling processing based on the network topology diagram and mycorrhizal network distribution map. Specifically, by transforming vessel transport and hyphal network transport into a dual problem of cooperative optimization, a set of coupled transport equations for vessel transport and mycorrhizal transfer is obtained.

[0072] Step S33: Perform carbon flux mode inversion processing based on the coupled transport equations and the digital gene tensor field. In this process, a regularized inversion algorithm based on variational principles is introduced to solve the coupled transport equations on the manifold space defined by the tensor field, thereby generating the dual-path carbon flux field.

[0073] Understandably, in this step, step S31 first extracts multi-scale features of the three-dimensional skeleton topology of trees from the digital gene tensor field, and separates the principal components describing the spatial orientation of the vascular bundles through tensor decomposition. Based on this, the base of the trunk is used as the initial element, and a vascular branching network that conforms to the growth law of trees is generated iteratively according to the fractal dimension. The fractal parameters are dynamically adjusted by the morphological features such as branch angles and curvature extracted from the tensor field. Then, for the physiological characteristics of xylem vessels, a fluid dynamic constraint based on the Hagen-Poiseuille law is introduced to model the vascular aperture distribution as a power-law function along the length of the vessel, the exponent of which is determined by the tree ring width and water conduction efficiency.

[0074] This step combines fractal geometry theory with plant physiological principles to reconstruct a continuous xylem network from roots to leaves. It not only simulates the spatial topology of xylem but also quantifies the differences in water conductivity among xylem of different diameter classes. A network topology map containing physical parameters such as xylem connectivity, pore size distribution, and water conductivity efficiency is generated, providing a physiologically realistic structural basis for subsequent carbon transport inversion.

[0075] The expression for generating a vascular branching network that conforms to the growth pattern of trees based on fractal dimension iteration is shown below:

[0076] ;

[0077] Among them, A n+1 A represents the branching state of the duct generated in the (n+1)th iteration. n D represents the branching state of the duct generated in the nth iteration. f The fractal dimension is determined by the morphological features extracted from the tensor field, θ represents the branch angle parameter, Φ represents the curvature parameter, and f is the generating function.

[0078] The power-law function for the duct orifice distribution is shown below:

[0079] ;

[0080] Where g(l) is the diameter of the vascular bundle at a distance l from the root, g0 is the reference diameter at the reference position l0, α is the power law exponent, R is the tree ring width, and E is the water conduction efficiency parameter.

[0081] The fluid dynamic constraints based on the Hagen-Poiseuille law are shown below:

[0082] ;

[0083] Where Q is the flow rate through the conduit, r is the conduit radius, π is pi, ΔP is the pressure difference, η is the liquid viscosity, and L is the conduit length.

[0084] Next, in step S32, a cooperative transport model based on constrained optimization theory is constructed to unify the pressure-driven flow within the conduits and the diffusion transport in the mycorrhizal network into a dual optimization problem. Specifically, the carbon flux generated by the water potential difference between the conduit network nodes is used as the main variable, and the diffusion flux driven by the carbon concentration difference between the mycorrhizal network nodes is used as the dual variable. An objective function is established to minimize the total transport dissipation energy of the system, where the constraint condition includes the nodal carbon conservation equation. In the solution process, the Lagrange multiplier method is used to transform the constrained optimization problem into a set of partial differential equations. The constitutive relationship between carbon flux and pressure gradient is derived for the conduit transport part based on the Hagen-Poiseuille law, while the relationship between carbon flux and concentration gradient is established for the mycorrhizal network part based on Fick's diffusion law. The two subsystems achieve cooperative transport under the minimum energy state through interface coupling terms.

[0085] This step transforms the "source-sink" relationship observed in biology into a mathematical dual optimization problem, which maintains the high efficiency of rapid vascular transport while utilizing the precise regulatory capabilities of mycorrhizal networks in local microenvironments, providing a mathematical model with clear physical meaning for subsequent carbon flux inversion.

[0086] The objective function to be minimized is shown below:

[0087] ;

[0088] in, This is the transpose of the carbon flux vector of the conduit network. It is the reciprocal of the conductivity tensor of the duct network, and is related to the duct pore size distribution and connectivity. Let be the carbon flux vector of the xylem network, representing the carbon transport rate through the xylem vessels. This is the transpose of the mycorrhizal network carbon flux vector. It is the reciprocal of the mycorrhizal network diffusion coefficient tensor, and is related to hyphal density and connectivity strength. Let be the carbon flux vector of the mycorrhizal network, representing the carbon transport rate through the hyphal network, d be the differential symbol, and V be the spatial volume of the entire forest system.

[0089] The nodal carbon conservation equation is as follows:

[0090] ;

[0091] in, Let S be the divergence operator, representing the divergence of a vector field. x For the carbon source and sink terms of the vascular network, λ represents the rate of carbon production or consumption per unit volume per unit time, where λ is the vascular-mycorrhizal network coupling coefficient, and C... x C represents the carbon concentration in the ductal network. m This represents the carbon concentration in the mycorrhizal network.

[0092] The partial differential equations are shown below:

[0093] ;

[0094] Where Φ is a Lagrange multiplier and Ψ is the computational domain, representing the spatial extent of the entire forest system.

[0095] Finally, the carbon flux mode inversion process in step S33 is as follows: First, the coupled transport equations are reconstructed into a variational problem on the manifold space. Then, by introducing an inversion framework based on Tikhonov regularization, an optimization objective function for carbon flux estimation is established on the Riemannian manifold defined by the digital gene tensor field. Specifically, a manifold metric tensor is constructed using the tree morphological gene features and ecological process information encoded in the tensor field. The residual terms of the vascular-mycorrhizal dual-path transport equations are combined with the smoothness constraints on the manifold to form a regularized least squares problem. In the solution process, a gradient descent algorithm based on the adjoint method is used. By constructing the adjoint operator of the transport operator, the gradient of the objective function with respect to the carbon flux field is efficiently calculated. Finally, the parallel transport theory on the manifold is combined to handle the geometric constraints between different tree individuals.

[0096] This step combines a physics-driven transport model with a data-driven manifold geometry, maintaining the physical consistency of carbon transport processes while fully utilizing the multi-scale ecological information contained in the digital gene tensor field. It obtains a spatially continuous carbon flux field distribution through variational inversion, quantifies the relative contributions of dominant vascular transport and recessive mycorrhizal translocation, and provides high-resolution flux benchmark data for carbon sink capacity assessment.

[0097] The regularization inversion objective function is shown below:

[0098] ;

[0099] in, The inversion objective function measures the degree of matching between model predictions and observed data. A is the forward transport operator, which maps carbon flux to the observation space. It is based on the coupled transport equations and includes flux components of dominant vascular transport and recessive mycorrhizal transfer. B is the observation data vector, which is the carbon flux-related feature from the digital gene tensor field. M is the manifold metric tensor, constructed from the digital gene tensor field, which encodes forest morphology and ecological process information. α is the regularization parameter, which balances the weights of the data fitting term and the regularization term. R is the regularization operator, which applies a smoothness constraint to the carbon flux field. Q is the carbon flux field vector.

[0100] Furthermore, step S4 includes steps S41, S42, and S43.

[0101] Step S41: The carbon storage dynamic system is constructed based on the dual-path carbon flux field. In this process, all flux components in the dual-path carbon flux field are transformed into a differential dynamic model with nonlinear coupling relationship. The differential dynamic model is analyzed using Lyapunov characteristic index spectrum to obtain the dynamic stability characteristics of carbon storage evolution.

[0102] Step S42: Based on the dynamic stability characteristics and mycorrhizal network distribution map, establish a hyphal channel connectivity model based on network percolation theory, and convert the mycorrhizal network topology into mycorrhizal buffer effect parameters based on the hyphal channel connectivity model.

[0103] Step S43: Phase transition boundary field generation processing is performed based on mycorrhizal buffer effect parameters and dual-path carbon flux field. Specifically, the carbon sink phase transition critical field with spatial explicit characteristics is constructed by solving the bifurcation surface of the carbon storage dynamic system in parameter space.

[0104] Understandably, in this step S41, the dominant vascular bundle transport flux and the recessive mycorrhizal transfer flux in the dual-path carbon flux field are first modeled as state variables. A nonlinear dynamic system describing the spatiotemporal evolution of carbon storage is constructed by introducing a differential equation framework based on the principle of mass conservation. The rate of change of carbon storage in each spatial unit is expressed as the difference between the input flux and the output flux. The carbon flux of the vascular bundle network is described by a convection-diffusion equation, while the carbon flux of the mycorrhizal network is expressed by a reaction-diffusion equation. The two subsystems are nonlinearly coupled through an interface carbon exchange term. Based on this, the Lyapunov characteristic index spectral analysis method is used to quantify the local stability of the carbon storage system near the equilibrium point by calculating the real parts of the eigenvalues ​​of the Jacobian matrix. The sign of the largest Lyapunov exponent determines the stability characteristics of the system.

[0105] Step S41 transforms the continuous carbon flux field into a dynamic system with clear physical meaning. The interaction mechanism between vascular transport and mycorrhizal transfer is accurately characterized by nonlinear coupling terms, providing a rigorous mathematical criterion for the long-term stability assessment of carbon sink function.

[0106] The differential equation framework based on the principle of mass conservation is shown below:

[0107] ;

[0108] Where C is the carbon storage state vector, t is the time variable, and Φ x and Φ m Φ represents the time scale of system evolution. x S is a nonlinear coupling function that describes the carbon exchange process between the xylem and mycorrhizal networks. x For source and sink terms, S represents the input process of photosynthesis. m This indicates the process of energy consumption during respiration.

[0109] The nonlinear coupling equations are shown below:

[0110] ;

[0111] Where k1 and k2 are the maximum exchange rate constants, representing the rate at which carbon exchange is controlled, and K1 and K2 are the half-saturation constants, representing the concentration at which carbon exchange reaches half-maximum rate.

[0112] Next, step S42 first identifies vulnerable regions of the system based on the Lyapunov exponent distribution in the dynamic stability characteristics, and abstracts the mycorrhizal network distribution map into a weighted graph model, where nodes represent mycorrhizal communities and edge weights are calculated through hyphal density and connection strength. On this basis, an improved seepage theory framework is introduced, and the phase transition behavior of randomly removing network edges and calculating the relative size of the largest connected component is simulated by Monte Carlo simulation. The removal probability is negatively correlated with the local Lyapunov exponent in the dynamic stability characteristics, so that regions with lower stability have a higher probability of edge failure. By quantifying the critical change in network connectivity near the seepage threshold, the topological robustness of the mycorrhizal network is transformed into a buffer effect parameter, which is defined as the standardized distance between the actual network connectivity and the critical seepage threshold.

[0113] Step S42 connects the abstract stability characteristics with the specific network topology through the percolation phase transition theory, realizing the quantitative characterization of the mycorrhizal network's ability to maintain the stability of the carbon sink system. It obtains parameter indicators that can quantify the buffering capacity of the mycorrhizal network when the carbon sink system is disturbed, providing key input parameters for the subsequent construction of the carbon sink phase transition critical field.

[0114] The formula for calculating the weighted graph model is as follows:

[0115] ;

[0116] Among them, W ij ρ represents the edge weight, indicating the connection strength between nodes. ij The mycelial density was obtained through molecular biological detection. ij For connection strength, λ is calculated based on the mycorrhizal network distribution map. i Let be the local Lyapunov exponent, derived from the dynamic stability feature, and exp be the exponential function.

[0117] The formula for the buffer effect parameter is as follows:

[0118] ;

[0119] Wherein, β is the mycorrhizal buffer effect parameter, which quantifies the network's resistance to disturbances, S(pz) is the actual network connectivity, which considers the stability characteristics of the network connectivity, S(pc) is the critical connectivity, which is the network connectivity at the seepage threshold, and S(0) is the complete network connectivity, which is the initial connectivity without removal.

[0120] Finally, step S43 first uses the mycorrhizal buffer effect parameter as a key control variable for the carbon storage dynamic system. An extended parameter space is constructed to couple the dual-path carbon flux field with the buffer parameter. Here, the critical surface of the carbon storage system's stability boundary is determined by tracking the parameter combinations when the system's Jacobian matrix eigenvalues ​​cross the imaginary axis. Specifically, the carbon flux field data of each spatial unit is combined with the corresponding mycorrhizal buffer parameter to form a multidimensional parameter vector. A prediction-correction algorithm is used to track the trajectory of the Hopf bifurcation point in the parameter space. Simultaneously, the coupling effect between adjacent spatial units is considered, and the geographical continuity of the bifurcation surface is ensured by introducing a spatial regularization term.

[0121] Step S43 combines abstract dynamic bifurcation theory with specific spatial ecological data, achieving a breakthrough from point-like bifurcation analysis to the construction of explicit spatial phase transition fields. This enables precise identification of critical conditions for abrupt changes in carbon sink function in different regions. It quantifies the stability boundary of each spatial unit under mycorrhizal network regulation, providing a spatially visualized early warning tool for carbon sink project risk management.

[0122] The space regularization term is shown below:

[0123] ;

[0124] Where R is the spatial regularization term, used to ensure the geographical continuity of the phase transition critical field and penalize drastic changes between adjacent spatial units, β c(x) is the critical buffer parameter field, which is the critical value of the mycorrhizal buffer effect at spatial location x. When the actual parameter exceeds this value, the system becomes unstable. x is the spatial coordinate.

[0125] Furthermore, step S5 includes steps S51, S52, and S53.

[0126] Step S51: Perform critical state network connectivity analysis processing based on the carbon sink phase transition field. Specifically, by extracting the critical point topology in the phase transition field, mapping it to a set of vulnerable nodes on the mycorrhizal network, and establishing a failure probability model of the carbon transport path based on network percolation theory, a network connectivity map considering phase transition risk is obtained.

[0127] Step S52: Optimize carbon redirection paths based on the network connectivity graph and mycorrhizal network distribution graph. This involves establishing a carbon particle multipath transport model based on random walk theory, calculating the minimum dissipation transport path of carbon along the hyphal network under the constraint of vulnerable nodes in the network, and obtaining the carbon redirection probability field.

[0128] Step S53: Generate a cross-scale carbon redistribution map based on the carbon redirection probability field and the carbon sink phase transition field. This is achieved by performing tensor convolution operations on the probability field and the phase transition field, and applying the renormalization group method at different spatial scales to obtain a carbon redistribution map based on biological network regulation.

[0129] It is understandable that step S51 identifies the topological features of critical points by calculating persistent homology, including the spatial distribution pattern and connectivity structure of critical points. Based on this, a mapping relationship between critical points and mycorrhizal network nodes is constructed, associating each critical point in the phase transition field with the nearest spatially located mycorrhizal network node, and assigning corresponding vulnerability weights based on the Lyapunov exponent values ​​of the critical points. Subsequently, a network failure model based on seepage theory is introduced, abstracting the mycorrhizal network into a weighted graph structure, where the node failure probability is negatively correlated with the stability parameters of the associated critical points. Monte Carlo simulations are used to calculate the phase transition behavior of the network's maximum connectivity component under different failure probabilities. Specifically, considering the biological characteristics of mycorrhizal networks in forestry carbon sink scenarios, the uniform failure assumption in the traditional seepage model is improved, taking into account the enhancing effect of redundant paths formed by hyphal connections between mycorrhizal network nodes on network robustness. By combining dynamical system stability theory with complex network theory, a cross-scale mapping from carbon sink functional stability to biological network connectivity is achieved.

[0130] The node failure model is shown below:

[0131] ;

[0132] Where, p is is the node failure probability, representing the probability that the node will lose connection under disturbance. s0 is a stability threshold parameter used to adjust the baseline value of the failure probability. i These are the stability parameters for the node.

[0133] The formula for calculating the network connectivity component is as follows:

[0134] ;

[0135] Where S(p) represents the relative size of the network connectivity components, N represents the total number of nodes in the network, and k represents the network node index. The indicator function is used when node k belongs to the maximum connected component. The value is 1 if it is active, and 0 otherwise.

[0136] Understandably, step S52 first integrates the node vulnerability information in the network connectivity graph with the topological structure of the mycorrhizal network distribution graph to construct a risk-weighted directed graph model. Node weights are determined by vulnerability indices, and edge weights are calculated using hyphal conduction efficiency. Based on this, a carbon particle transport model based on random walk theory is introduced, modeling the carbon migration process in the mycorrhizal network as a continuous-time random walk. The transfer probability depends not only on the conduction capacity of hyphal connections but also on the node vulnerability. The steady-state probability distribution of carbon particles under risk constraints is obtained by solving the Fokker-Planck equation with absorbing boundary conditions. Specifically, considering the biological characteristics of mycorrhizal networks in forestry carbon sink scenarios, a minimum dissipation principle based on metabolic cost is introduced into the random walk model. The energy functional extrema of the carbon transport path are solved using variational methods to obtain the optimal path set that minimizes system dissipation. Step S52 combines network risk constraints with the physical dissipation principle, achieving risk-aware optimization of carbon transport paths, considering both the topological characteristics of the mycorrhizal network and stability risk factors.

[0137] The formula for calculating the risk-weighted edge weights is as follows:

[0138] ;

[0139] Where, ε ij For risk-weighted edge weights, c ij v represents the hyphal conduction efficiency. i和 v j All are indicators of node vulnerability, d ij l is the length of the hyphal path. o The feature length scale.

[0140] The formula for calculating the energy functional is as follows:

[0141] ;

[0142] in, It is an energy functional that describes the total energy dissipated during carbon transport. ρ(x) represents the absolute value of the carbon flux field, σ(x) represents the electrical conductivity field, V(x) represents the potential energy field, and ρ(x) represents the carbon density field.

[0143] It is understandable that step S53 first aligns the carbon redirection probability field and the carbon sink phase transition field using multimodal data, ensuring geometric and topological consistency between the two fields by establishing a unified spatial reference framework. Based on this, a cross-scale information fusion method using tensor convolution is employed to couple the path selection information in the carbon redirection probability field with the stability boundary information in the carbon sink phase transition field in a high-order manner, enabling it to simultaneously capture local path features and global phase transition modes. Specifically, a renormalization group method is introduced to handle multi-scale effects. By constructing a scale transformation operator, the convolution results are recursively coarsened, redefining the effective carbon transport parameters at each scale, thereby establishing a carbon redistribution pattern from the microscopic mycorrhizal network to the macroscopic forest stand scale. Considering the biological characteristics of forestry carbon sink scenarios, the heterogeneity of the mycorrhizal network is preserved during the renormalization process, avoiding the loss of biological details caused by scale transformation in traditional methods.

[0144] By fusing path optimization information from the probability field with stability information from the phase transition field through rigorous mathematical transformation, a multi-scale consistent characterization of the carbon redistribution process is achieved.

[0145] The multimodal data alignment formula is shown below:

[0146] ;

[0147] Where T1 is the transformed carbon redirection probability field, R1 is the rotation matrix used for spatial transformation of coordinate system one, P1 is the original carbon redirection probability field, representing the carbon transport probability distribution in the mycorrhizal network, and t1 is the translation vector used for spatial position correction.

[0148] The formula for calculating tensor convolution is as follows:

[0149] ;

[0150] in, For the resulting convolution field, Represents the carbon redirection probability field. Here, x, y, u, and v are the convolution kernel function, and x, y, u, and v are spatial coordinate variables representing the position information in the field.

[0151] Furthermore, step S6 includes steps S61, S62, and S63.

[0152] Step S61: Perform carbon storage measurement manifold construction processing based on the carbon redistribution map, wherein the carbon distribution at the start and end of the project is expressed as a probability measure in Wasserstein space, and the connection structure on the manifold is defined using the carbon redistribution map to obtain the carbon storage measurement manifold.

[0153] Step S62: Perform entropy regularization optimal transport calculation based on the carbon storage measurement manifold to obtain a regularized optimal transport plan;

[0154] Step S63: Perform lifecycle uncertainty propagation processing based on the regularized optimal transport plan. This involves performing tensor convolution operations between the optimal transport plan and the carbon redistribution map, while maintaining carbon flux conservation constraints in the lifecycle dimension, to obtain the full lifecycle carbon sink estimate adjusted by entropy optimization.

[0155] Understandably, in this step, step S61 first characterizes the carbon distribution at the beginning and end of the project as a probability measure in Wasserstein space, and establishes a basic metric structure by calculating the difference in the second moment between the carbon storage distributions of the two periods. On this basis, a connection construction method for the carbon redistribution map is introduced, which transforms the carbon redirection path information encoded in the map into affine connection coefficients on the manifold. Specifically, the parallel movement rules at each point of the manifold are defined by calculating the covariant derivative of the carbon flux path. In particular, considering the non-uniform distribution characteristics of carbon storage in the forestry carbon sink scenario, the Ben-Schuler regularization method in optimal transport theory is used to geometrically smooth the metric manifold, so that the manifold structure can reflect both the macroscopic distribution characteristics of carbon storage and retain the microscopic transport path information under the regulation of mycorrhizal networks.

[0156] Step S61 transforms the traditional static comparison of carbon reserves into dynamic geometric analysis. By using the geometric structure of the Wasserstein manifold to naturally characterize the evolution trajectory of carbon reserve distribution during the project period, a carbon reserve measurement manifold with an inherent interconnected structure is constructed, providing a differential geometric framework that maintains the carbon redistribution constraint for subsequent entropy regularization optimal transport calculations.

[0157] The linkage correction under the carbon redistribution constraint is shown below:

[0158] ;

[0159] in, Let be the affine connection coefficient, describing the rules governing parallel vector translation on a manifold. For the carbon redistribution tensor, geometric constraints are derived from the carbon redistribution map. The coupling strength parameter controls the degree to which carbon redistribution affects the geometry. The corrected connection coefficient includes carbon redistribution constraints.

[0160] The Ben-Schuller regularization terms are as follows:

[0161] ;

[0162] in, The Ben-Schuler regularization term is used to smooth the measure manifold, where M is the fundamental manifold and represents the spatial domain of the forestry project. The gradient of the logarithm of the probability density describes the non-uniformity of carbon distribution, and μ is a probability measure representing the carbon storage distribution at the start of the project.

[0163] Next, step S62 first constructs a Wasserstein distance framework on the carbon storage measurement manifold, using the carbon distribution at the beginning and end of the project as the source and target measures, respectively. By introducing a regularization term based on relative entropy, the traditional optimal transport problem is transformed into a convex optimization problem with entropy constraints. Specifically, efficient computation is achieved on the manifold using the Sinkhorn iterative algorithm. The entropy regularization parameter is dynamically adjusted according to the path uncertainty in the carbon redistribution map, and the connection structure on the manifold maintains the consistency between the transport path and the carbon redistribution constraints. Step S62 overcomes the problems of noise sensitivity and excessive computational complexity of traditional optimal transport through entropy regularization, while maintaining the physical rationality of the carbon redistribution path.

[0164] The convex optimization with entropy constraints is shown below:

[0165] ;

[0166] in, M represents the set of all coupling measures. o The manifold represents the carbon storage measurement, and the spatial distribution of carbon is represented. This represents the optimal transmission plan after regularization. Let ε represent the relative entropy, and ε be the regularization parameter.

[0167] Finally, step S63, based on the dynamic integral method of tensor convolution, performs multidimensional convolution operations on the optimal transport plan and the carbon redistribution map. The convolution kernel function is adaptively designed according to the path weights and temporal characteristics in the carbon redistribution map, ensuring that the convolution process can capture the cross-scale migration patterns of carbon in mycorrhizal networks and vascular systems. Specifically, considering the dynamic characteristics of the entire life cycle of forestry carbon sinks, a carbon flux conservation constraint equation is constructed, embedding a mass balance condition into the convolution integral. The Lagrange multiplier method is used to transform flux conservation into a hard constraint in the optimization problem, thus ensuring that changes in carbon storage conform to the law of conservation of matter in the ecosystem. Regarding uncertainty propagation, the uncertainty information contained in the entropy term in the regularized optimal transport plan is utilized. By simulating the cumulative error propagation of carbon flux over the life cycle, a Monte Carlo integral is finally used to perform a probability-weighted average of the convolution results. Step S63 combines static optimal transport theory with the dynamic carbon cycle process, achieving a natural fusion of carbon redistribution paths and the time dimension through tensor convolution, while automatically quantifying uncertainties in the transport process using entropy regularization.

[0168] The formula for tensor convolution is as follows:

[0169] ;

[0170] Where V(x,t) is the spatiotemporal carbon sink valuation field, that is, the carbon sink value at location x and time t, which is the result of convolution operation, and Z is the project cycle length. The kernel function for the optimal transmission plan represents the time delay from position y. The probability density of subsequent transmission to x. This represents a carbon redistribution map. The spatial computing domain is the spatial extent of the entire forest area.

[0171] The carbon flux conservation constraint equation is shown below:

[0172] ;

[0173] Where C(x,t) is the carbon concentration field, the carbon storage density at position x and time t, J(x,t) is the carbon flux field, and S(x,t) is the carbon source and sink term.

[0174] Finally, the formula for calculating the full life-cycle carbon sink valuation is as follows:

[0175] ;

[0176] in, Valuation of carbon sinks over their entire life cycle. Valuation variance is a measure of the uncertainty in carbon sink valuation. As an uncertainty adjustment factor, This represents the risk aversion coefficient.

[0177] Example 2:

[0178] like Figure 2 As shown, this embodiment provides a digital twin-based full life-cycle measurement system for forestry carbon sink projects. See [link to documentation]. Figure 2 The system includes an acquisition unit 701, a processing unit 702, an inversion unit 703, an analysis unit 704, a deduction unit 705, and an estimation unit 706.

[0179] The acquisition unit 701 is used to acquire multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, distribution map of mycorrhizal network and time series data of meteorology.

[0180] Processing unit 702 is used to couple the multi-source heterogeneous basic data with forest digital genes and ecological process tensors, wherein feature extraction is performed on the multi-source heterogeneous basic data and the extracted features are mapped to generate a digital gene tensor field.

[0181] Inversion unit 703 is used to perform carbon transport inversion processing driven by vascular bundle network on digital gene tensor field, wherein by decomposing digital gene tensor field into carbon flux modes transported along vascular bundle network, a dual-path carbon flux field including dominant vascular transport and recessive mycorrhizal transfer is obtained.

[0182] The analysis unit 704 is used to perform carbon sink phase transition critical point analysis processing on the dual-path carbon flux field. In this process, by identifying the critical conditions under which the carbon storage state undergoes a sudden change and quantifying the influence of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated.

[0183] The deduction unit 705 is used to perform cross-scale carbon redistribution path deduction processing based on the carbon sink phase transition field to obtain a carbon redistribution map based on biological network regulation.

[0184] The estimation unit 706 is used to perform full life cycle carbon sink entropy optimization measurement based on the carbon redistribution map to obtain the optimized and adjusted full life cycle carbon sink estimate.

[0185] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0186] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0187] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for measuring the entire life cycle of forestry carbon sink projects based on digital twins, characterized in that, include: Acquire multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, distribution map of mycorrhizal network and time series data of meteorology. The multi-source heterogeneous basic data is coupled with forest digital genes and ecological process tensors. In this process, features are extracted from the multi-source heterogeneous basic data and the extracted features are mapped to generate a digital gene tensor field. The digital gene tensor field is subjected to carbon transport inversion driven by vascular bundle network. Specifically, by decomposing the digital gene tensor field into carbon flux modes that are transported along the vascular bundle network, a dual-path carbon flux field containing dominant vascular transport and recessive mycorrhizal transfer is obtained. The dual-path carbon flux field is subjected to carbon sink phase transition critical point analysis processing. Specifically, by identifying the critical conditions for abrupt changes in carbon storage state and quantifying the influence of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated. Based on the carbon sink phase transition field, cross-scale carbon redistribution path deduction was performed to obtain a carbon redistribution map based on biological network regulation. Based on the carbon redistribution map, the full life cycle carbon sink entropy optimization measurement process is performed to obtain the optimized and adjusted full life cycle carbon sink estimate. The coupling processing of the multi-source heterogeneous basic data with forest digital genes and ecological process tensors includes: The morphological gene vector extraction process is performed on the three-dimensional skeleton topological structure data of the forest trees. Specifically, by performing topological persistence analysis on the point cloud skeleton, the trunk curvature data, branch fractal dimension data, and canopy filling data are extracted and encoded into high-dimensional feature vectors to obtain the forest tree morphological gene vectors. Based on the aforementioned forest tree morphological gene vectors and mycorrhizal network distribution map, underground carbon transport channels are coupled. Specifically, by converting the mycorrhizal network map into a weighted directed graph structure, the morphological gene vectors are projected onto the network spectral domain space using the Laplace eigenmap in graph theory to obtain forest tree-mycorrhizal joint gene units. Tensor field construction processing is performed based on the aforementioned forest tree-mycorrhizal joint gene units and carbon dioxide flux time series data. Specifically, a digital gene tensor field with multi-resolution characteristics is constructed by performing tensor product operation on the singular spectrum of the joint gene units and the flux time series.

2. The method for full life-cycle measurement of forestry carbon sink projects based on digital twins according to claim 1, characterized in that, Carbon transport inversion processing driven by duct bundle networks is performed on the digital gene tensor field, including: The xylem vessel network is reconstructed based on the digital gene tensor field. Specifically, the vessel bundle connection relationship contained in the three-dimensional skeleton topology of the forest in the tensor field is analyzed, and a continuous vessel network model from the root system to the canopy is reconstructed based on fractal geometry theory to obtain a network topology map containing vessel pore size distribution and connectivity characteristics. Based on the network topology and mycorrhizal network distribution map, a dual-path carbon transport coupling process is performed. Specifically, by transforming vessel transport and hyphal network transport into a dual problem of synergistic optimization, a set of coupled transport equations for vessel transport and mycorrhizal transfer is obtained. Carbon flux mode inversion is performed based on the coupled transport equations and the digital gene tensor field. Specifically, a regularized inversion algorithm based on variational principles is introduced to solve the coupled transport equations on the manifold space defined by the tensor field, thereby generating the dual-path carbon flux field.

3. The method for full life-cycle measurement of forestry carbon sink projects based on digital twins according to claim 1, characterized in that, The dual-path carbon flux field is subjected to carbon sink phase transition critical point analysis, including: The carbon storage dynamic system is constructed based on the dual-path carbon flux field. In this process, all flux components in the dual-path carbon flux field are transformed into a differential dynamic model with nonlinear coupling relationship. The dynamic stability characteristics of carbon storage evolution are obtained by analyzing the differential dynamic model using the Lyapunov characteristic index spectrum. Based on the dynamic stability characteristics and mycorrhizal network distribution map, a hyphal channel connectivity model based on network percolation theory is established, and the mycorrhizal network topology is transformed into mycorrhizal buffer effect parameters based on the hyphal channel connectivity model. Phase transition boundary field generation is performed based on mycorrhizal buffer effect parameters and dual-path carbon flux field. Specifically, a carbon sink phase transition critical field with spatial explicit characteristics is constructed by solving the bifurcation surface of the carbon storage dynamic system in parameter space.

4. The method for full life-cycle measurement of forestry carbon sink projects based on digital twins according to claim 1, characterized in that, Based on the aforementioned carbon sink phase transition field, cross-scale carbon redistribution path deduction processing is performed, including: The critical state network connectivity analysis is performed based on the carbon sink phase transition field. Specifically, the critical point topology in the phase transition field is extracted and mapped to a set of vulnerable nodes on the mycorrhizal network. A failure probability model of the carbon transport path is established based on network percolation theory to obtain a network connectivity map considering the phase transition risk. Carbon redirection path optimization is performed based on the network connectivity graph and mycorrhizal network distribution graph. Specifically, a carbon particle multipath transport model based on random walk theory is established, and the minimum dissipation transport path of carbon along the hyphal network is calculated under the constraint of vulnerable nodes in the network, thus obtaining the carbon redirection probability field. A cross-scale carbon redistribution map is generated based on the carbon redirection probability field and the carbon sink phase transition field. Specifically, by performing tensor convolution operations on the probability field and the phase transition field, and applying the renormalization group method at different spatial scales, a carbon redistribution map based on biological network regulation is obtained.

5. A full life-cycle measurement system for forestry carbon sequestration projects based on digital twins, characterized in that, include: The acquisition unit is used to acquire multi-source heterogeneous basic data of the target forest area to be calculated. The multi-source heterogeneous basic data includes three-dimensional skeleton topology data of trees, time series data of carbon dioxide flux, distribution map of mycorrhizal network and time series data of meteorology. The processing unit is used to couple the multi-source heterogeneous basic data with forest digital genes and ecological process tensors. Specifically, by extracting features from the multi-source heterogeneous basic data and mapping the extracted features, a digital gene tensor field is generated. The inversion unit is used to perform carbon transport inversion processing driven by vascular bundle network on the digital gene tensor field. In this process, by decomposing the digital gene tensor field into carbon flux modes that are transported along the vascular bundle network, a dual-path carbon flux field containing dominant vascular transport and recessive mycorrhizal transfer is obtained. The analysis unit is used to perform carbon sink phase transition critical point analysis processing on the dual-path carbon flux field. Specifically, by identifying the critical conditions for abrupt changes in carbon storage state and quantifying the influence of mycorrhizal network buffering on the phase transition threshold, a carbon sink phase transition field with stable boundary constraints is generated. The deduction unit is used to perform cross-scale carbon redistribution path deduction processing based on the carbon sink phase transition field to obtain a carbon redistribution map based on biological network regulation. The estimation unit is used to perform full life cycle carbon sink entropy optimization measurement based on the carbon redistribution map to obtain the optimized and adjusted full life cycle carbon sink estimate. The processing unit includes: The first processing subunit is used to extract morphological gene vectors from the three-dimensional skeleton topological structure data of the forest trees. Specifically, by performing topological persistence analysis on the point cloud skeleton, the trunk curvature data, branch fractal dimension data, and canopy filling data are extracted and encoded into high-dimensional feature vectors to obtain forest tree morphological gene vectors. The second processing subunit is used to perform underground carbon transport channel coupling processing based on the forest tree morphological gene vector and mycorrhizal network distribution map. In this process, the mycorrhizal network map is transformed into a weighted directed graph structure, and the morphological gene vector is projected onto the network spectral domain space using the Laplace eigenmap in graph theory to obtain the forest tree-mycorrhizal joint gene unit. The third processing subunit is used to construct a tensor field based on the forest tree-mycorrhizal joint gene unit and carbon dioxide flux time series data. Specifically, a digital gene tensor field with multi-resolution characteristics is constructed by performing tensor product operation between the joint gene unit and the singular spectrum of the flux time series.

6. The forestry carbon sequestration project lifecycle measurement system based on digital twins according to claim 5, characterized in that, The inversion unit includes: The first inversion subunit is used to reconstruct the xylem vessel network based on the digital gene tensor field. Specifically, by analyzing the vessel bundle connection relationship contained in the three-dimensional skeleton topology of the forest in the tensor field, and reconstructing the continuous vessel network model from the root system to the canopy based on fractal geometry theory, a network topology map containing vessel pore size distribution and connectivity characteristics is obtained. The second inversion subunit is used to perform dual-path carbon transport coupling processing based on the network topology map and mycorrhizal network distribution map. In this process, by transforming vessel transport and hyphal network transport into a dual problem of synergistic optimization, a set of coupled transport equations for vessel transport and mycorrhizal transfer is obtained. The third inversion subunit is used to perform carbon flux mode inversion processing based on the coupled transport equations and the digital gene tensor field. The coupled transport equations are solved on the manifold space defined by the tensor field by introducing a regularized inversion algorithm based on the variational principle to generate the dual-path carbon flux field.

7. The forestry carbon sequestration project lifecycle measurement system based on digital twins according to claim 5, characterized in that, The parsing unit includes: The first analytical subunit is used to construct a dynamic carbon storage system based on the dual-path carbon flux field. In this process, all flux components in the dual-path carbon flux field are transformed into a differential dynamic model with nonlinear coupling, and the differential dynamic model is analyzed using the Lyapunov characteristic index spectrum to obtain the dynamic stability characteristics of carbon storage evolution. The second analytical subunit is used to establish a hyphal channel connectivity model based on network percolation theory based on the dynamic stability characteristics and mycorrhizal network distribution map, and to convert the mycorrhizal network topology into mycorrhizal buffer effect parameters based on the hyphal channel connectivity model. The third analytical subunit is used to generate phase transition boundary fields based on mycorrhizal buffer effect parameters and dual-path carbon flux fields. Specifically, by solving the bifurcation surface of the carbon storage dynamic system in parameter space, a carbon sink phase transition critical field with spatial explicit characteristics is constructed.

8. The forestry carbon sequestration project lifecycle measurement system based on digital twins according to claim 5, characterized in that, The deduction unit includes: The first deduction subunit is used to perform critical state network connectivity analysis based on the carbon sink phase transition field. Specifically, by extracting the critical point topology in the phase transition field, it is mapped to a set of vulnerable nodes on the mycorrhizal network. Based on the network percolation theory, a failure probability model of the carbon transport path is established to obtain a network connectivity map considering the phase transition risk. The second deduction subunit is used to perform carbon redirection path optimization based on the network connectivity map and mycorrhizal network distribution map. In this process, a carbon particle multipath transport model based on random walk theory is established, and the minimum dissipation transport path of carbon along the hyphal network is calculated under the constraint of vulnerable nodes in the network, so as to obtain the carbon redirection probability field. The third deduction subunit is used to generate a cross-scale carbon redistribution map based on the carbon redirection probability field and the carbon sink phase transition field. The carbon redistribution map based on biological network regulation is obtained by performing tensor convolution operation on the probability field and the phase transition field and applying the renormalization group method at different spatial scales.

Images