Regional GPP assimilation inversion system fusing ground-based multi-source satellite chlorophyll fluorescence

Abstract

The invention discloses a regional GPP assimilation inversion system fusing foundation-multi-source satellite chlorophyll fluorescence, and belongs to the technical field of terrestrial ecosystem carbon cycle monitoring. The signal unmixing and simulation generation module is used for unmixing the total fluorescence of the mixed pixels into pure chlorophyll fluorescence (SIF) and driving an ecological model to generate SIF and GPP simulation values based on environmental stress data; the state component decoupling module is used for decoupling the total GPP analog value into a light response fast component, a light response slow component and a light response memory component; the state evolution prediction module is used for predicting dynamic evolution of the slow component and the memory component; and the self-adaptive assimilation inversion module is used for performing inversion on the region GPP by taking the slow component and the memory component as state variables, taking the SIF observation value as observation data and combining a prediction result of the state evolution prediction module. According to the method, systematic deviation caused by model simplification or data singleness in a traditional method is overcome.

Description

Technical Field

[0001] This invention relates to the field of terrestrial ecosystem carbon cycle monitoring, specifically a regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence. Background Technology

[0002] Existing technologies for estimating total primary productivity (TPP) over large regions face several technical bottlenecks, including the physiological time lag of vegetation photosynthesis in response to the environment, nonlinear abrupt changes in ecosystems under stress, and nonlinear mixing of multiple vegetation functional signals within remote sensing pixels. Furthermore, at the data source level, existing technologies face a fundamental problem: numerous SIF remote sensing data sources are currently available, including GOSAT, GOME-2, OCO-2, and TROPOMI; however, differences in the observation instruments, inversion algorithms, time spans, spectral and spatial resolutions, and revisit cycles among these satellites lead to inconsistent quality of various SIF data products. This poses a significant challenge to the effective fusion and collaborative application of multi-source data, and these issues urgently need to be addressed.

[0003] To address the aforementioned issues, a novel assimilation and inversion method is proposed. This method first models the dynamic evolution of the ecosystem state, then decouples the complex time response of photosynthesis, and then deconstructs the mixed signals from remote sensing observations. Finally, it integrates multi-source information within an adaptive framework to achieve accurate inversion.

[0004] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence to solve the problems mentioned in the background art.

[0006] The technical solution of the present invention includes: The observation data acquisition module is used to acquire data on solar-induced chlorophyll fluorescence, photosynthetically active radiation, high-resolution land cover, and environmental stress from ground-based and remote sensing observations. The signal demixing and simulation generation module is used to demix the total fluorescence of mixed pixels into pure chlorophyll fluorescence (SIF) based on the vegetation functional coverage determined by high-resolution land cover data, and drive the ecological model to generate SIF and GPP simulation values ​​based on environmental stress data. The state component decoupling module is used to decouple the total GPP into fast optical response components, slow optical response components, and memory components. The state evolution prediction module is used to predict the dynamic evolution of the slow component and the memory component based on a preset ecosystem state critical transition model. The adaptive assimilation and inversion module is used to invert the regional GPP by using the slow component and memory component as state variables, the SIF observations as observation data, and combining the prediction results of the state evolution prediction module.

[0007] Preferably, the signal demixing and analog generation module is specifically used for: By applying a pre-defined ecological mechanism-constrained mixed pixel unmixing model, combined with vegetation functional coverage, a pre-defined canopy radiative transfer correction operator, and ecological interaction terms, the pure chlorophyll fluorescence SIF is calculated. Using a pre-defined physiologically constrained nonlinear transformation operator, a model was established to transform the fluorescence intensity (SIF) of pure chlorophyll into the maximum carboxylation rate (V). cmax25 The quantitative relationship.

[0008] Preferably, the ecological interaction term is modeled as a function with the coverage, leaf area index and exclusive competition coefficient between different vegetation functional types as inputs, and is used to describe the nonlinear effects caused by resource competition or shading between different vegetation functional types.

[0009] Preferably, the physiological constraint nonlinear transformation operator includes a nonlinear adjustment term, which is related to temperature and water vapor pressure difference in the environmental stress data and is used to characterize the regulation of the electron transport chain by environmental stress.

[0010] Preferably, the state component decoupling module is specifically used for: Based on the photosynthetically active radiation obtained by the observation data acquisition module, the fast component of the light response is determined. By applying a pre-defined temporal convolution model and integrating the historical environmental stress, the memory component is determined.

[0011] Preferably, the temporal convolution model employs a double-exponential memory kernel function to capture the composite memory characteristics of rapid vegetation recovery and slow adaptation.

[0012] Preferably, the critical transition model of the ecosystem state is a nonlinear stochastic differential equation based on bifurcation theory, used to describe the nonlinear response of the ecosystem under environmental stress.

[0013] Preferably, the adaptive assimilation and inversion module is specifically used for: Construct an assimilation framework with slow components and memory components as state variables; The state transition equations of the assimilation framework are based on the state evolution prediction module. The clean SIF observations provided by the signal demixing and simulation generation module are used as the observation inputs to the assimilation framework, and the state variables and model parameters are optimized through data assimilation.

[0014] Preferably, the adaptive assimilation and inversion module is further used for: Based on the dominant vegetation functional type within the pixel, the corresponding parameters are retrieved from the preset physiological parameter library; Based on the retrieved parameters, the constraint terms in the assimilation cost function are dynamically adjusted to ensure that the model evolution conforms to the physiological laws of specific vegetation functional types.

[0015] This invention provides an improved regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence, offering the following improvements and advantages compared to existing technologies: 1. This scheme achieves dynamic tracking and accurate inversion of the carbon cycle process in the ecosystem by constructing a complete closed-loop system consisting of observation data acquisition, signal processing, state decoupling, dynamic prediction, and adaptive assimilation. The overall advantage lies in the fusion of physical mechanism models and multi-source remote sensing observation data within a unified assimilation framework, which makes the model predictions constrained by real-time observations, while the interpretation of observation data is guided by the physical model, thereby overcoming the systematic bias caused by traditional methods due to model simplification or data uniformity. 2. This scheme introduces a hybrid pixel unmixing model based on ecological mechanism constraints, especially the defined ecological interaction term, to achieve a more accurate description of this physical process. The ecological interaction term of this scheme can quantify the nonlinear attenuation of herbaceous fluorescence signal caused by this shading, thereby calculating a more realistic pure chlorophyll fluorescence SIF value, which is impossible to achieve by traditional linear models. 3. This scheme establishes a range from pure chlorophyll fluorescence SIF to the maximum carboxylation rate V through this nonlinear adjustment term. cmax25 The quantitative relationship enables the final inverted GPP to truly reflect the physiological state of vegetation under different stress conditions, greatly improving the reliability of the inversion results under extreme climate events. 4. The adaptive assimilation and inversion module of this scheme solves the problem of the one-size-fits-all approach of traditional models by introducing a parameterization scheme based on vegetation functional types. The system can retrieve specific parameters from a preset physiological parameter library according to the dominant vegetation functional type in a pixel, such as C3 and C4 plants, and dynamically adjust the constraint terms in the assimilation cost function. This means that the behavior of the model itself will be adaptively adjusted according to the physiological characteristics of its simulated objects. For example, it strengthens the adaptive constraint of high light intensity for C4 crops and enhances the sensitivity constraint of temperature for C3 forests. This strategy of explicitly embedding macro-ecological knowledge into the assimilation process in mathematical form ensures the seamless transition of inversion results between different ecosystem types and the authenticity of internal details. Its accuracy and reliability are far superior to traditional models that use general parameters. 5. This scheme establishes a transformation from pure chlorophyll fluorescence (SIF) to the maximum carboxylation rate (V) using a physiologically constrained nonlinear transformation operator. cmax25The quantitative relationship of dynamic regulation under environmental stress; this operator enables the assimilation system to effectively constrain the true physiological state of vegetation using fluorescence observations, thereby significantly improving the reliability of GPP inversion results under extreme climate events such as drought and high temperature; the adaptive assimilation inversion module of this scheme can retrieve specific parameters based on the dominant vegetation functional type within the pixel, and embed macro-ecological knowledge into the assimilation process in mathematical form, ensuring the authenticity of the inversion results; this invention transforms the originally cumbersome and time-consuming standardization work of multi-source satellite data into an automated process by integrating an automated data preprocessing process. Attached Figure Description

[0016] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence according to the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0018] Example 1: Please see Figure 1 This invention provides a regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence, comprising: The observation data acquisition module is used to acquire high-resolution land cover data, including solar-induced chlorophyll fluorescence, photosynthetically active radiation, and environmental stress data, from ground-based and remote sensing observations. Its function is to provide the inversion system with standardized, preprocessed multi-source input data. The module integrates an automated preprocessing workflow to address the issue of inconsistent SIF data quality from different satellite sources, specifically including: The data spatial aggregation unit serves as the entry point for data processing. This unit performs quality control and regional cropping on the raw satellite data based on multiple judgment criteria. Then, the filtered data is spatially aggregated into a standard grid of 0.05 degrees and outputs a celestial-scale nc format data file. The data time aggregation unit can further aggregate standard datasets at the daily scale into monthly and yearly scales to facilitate subsequent data mining and research. The multi-source data joint unit matches and combines data from different sources that have undergone spatiotemporal standardization to form a unified dataset, facilitating further research and analysis.

[0019] The signal demixing and simulation generation module is used to demix the total fluorescence of mixed pixels into pure chlorophyll fluorescence (SIF) based on the vegetation functional coverage determined by high-resolution land cover data, and drive the ecological model to generate SIF and GPP simulation values ​​based on environmental stress data. The state component decoupling module is used to decouple the total GPP into fast optical response components, slow optical response components, and memory components. The state evolution prediction module is used to predict the dynamic evolution of the slow component and the memory component based on a preset ecosystem state critical transition model. The adaptive assimilation and inversion module is used to invert the regional GPP by using the slow component and memory component as state variables, the SIF observations as observation data, and combining the prediction results of the state evolution prediction module. This embodiment provides a regional GPP assimilation and inversion system that integrates ground-based and multi-source satellite chlorophyll fluorescence; the system integrates multiple functional modules to form a complete technical closed loop from raw remote sensing data input to final GPP product output; The system includes an observation data acquisition module. The purpose of this module is to provide the necessary, multi-source input data for the entire inversion system. In this embodiment, the module periodically acquires four types of data by accessing a data distribution service system and a meteorological reanalysis database: The first type is sunlight-induced chlorophyll fluorescence (total fluorescence of mixed pixels) from remote sensing observations, which is the core observation and can be obtained from hyperspectral satellite sensors such as TanSat, OCO-2, and TROPOMI; the second type is photosynthetically active radiation, reflecting the energy input of photosynthesis, which can be obtained from geostationary or polar-orbiting satellite products; the third type is high-resolution land cover data, used to identify the type and spatial distribution of surface vegetation, which can be obtained from high-resolution imagers such as Landsat and Sentinel; the fourth type is environmental stress data, mainly including surface temperature and atmospheric vapor pressure difference, used to quantify the degree of environmental limitation on photosynthesis, usually obtained from meteorological reanalysis datasets such as ERA5. The signal demixing and simulation generation module aims to convert the coarse mixed signals observed by remote sensing into SIF observation values ​​with clear physiological significance that can be directly utilized by the assimilation system. It receives high-resolution land cover data provided by the observation data acquisition module and determines the coverage of different vegetation functional types within each remote sensing pixel. Subsequently, it deconstructs the total fluorescence signal of the mixed pixel and separates the pure chlorophyll fluorescence SIF signal belonging to a specific PFT. The state component decoupling module aims to decompose the complex process of total vegetation population (GPP) into multiple components with different dynamic characteristics according to its different time scales of response to the environment. In this embodiment, the total GPP is decoupled into three parts: a fast light response component, a slow light response component, and a memory component. The fast light response component refers to the part of the GPP directly driven by current light conditions, such as minute- to hourly PAR changes. The slow light response component refers to the part of the GPP related to vegetation physiological states, such as photosynthetic capacity and leaf nitrogen content, with a change cycle of several days to several weeks. The memory component refers to the part of the GPP reflecting the cumulative or residual effects of historical environmental stresses, with a change scale of several weeks to several months. The State Evolution Prediction Module aims to predict the future dynamics of non-transient response components in GPP based on ecological mechanisms. It employs a pre-defined ecosystem state critical transition model to perform the prediction. The model specifically describes the dynamic evolution process of the decoupled slow and memory components, and can simulate the stability, change, and even mutation behavior of the ecosystem under internal and external driving forces. The adaptive assimilation and inversion module is the core computational unit of the entire system. Its purpose is to integrate model predictions with actual observations to achieve optimal estimation of the regional GPP. A data assimilation framework is constructed, in which the slow component and memory component, determined by the state component decoupling module, are defined as the system's state variables, representing the core physiological states that need to be tracked and optimized. The SIF observations produced by the signal demixing and simulation generation module are introduced into the assimilation framework as external observation data. The module combines the next-time state prediction provided by the state evolution prediction module with the current-time pure SIF observations provided by the signal demixing and simulation generation module, and continuously adjusts the state variables through optimization algorithms such as Kalman filtering or variational assimilation to achieve accurate inversion of the total GPP. By organically combining the above modules, this system constructs a complete framework that can dynamically track and predict the photosynthetic physiological state of vegetation; it solves the problem that existing technologies cannot effectively handle physiological time delays, nonlinear mutations and signal mixing. By decoupling GPP at multiple scales and fusing multi-source information within the assimilation framework, it achieves higher-precision inversion of the spatiotemporal dynamic changes of regional GPP, providing more reliable data support for carbon cycle research and ecosystem health monitoring. Compared to existing static or empirical GPP estimation methods, this approach constructs a complete closed-loop system consisting of observational data acquisition, signal processing, state decoupling, dynamic prediction, and adaptive assimilation. This enables dynamic tracking and accurate inversion of ecosystem carbon cycle processes. The overall advantage lies in fusing physical mechanism models with multi-source remote sensing observation data within a unified assimilation framework. This allows model predictions to be constrained by real-time observations, while the interpretation of observational data is guided by the physical model. This overcomes the systematic biases caused by model simplification or data uniformity in traditional methods.

[0020] Example 2 The signal demixing and analog generation module is specifically used for: By applying a pre-defined ecological mechanism-constrained mixed pixel unmixing model, combined with vegetation functional coverage, a pre-defined canopy radiative transfer correction operator, and ecological interaction terms, the pure chlorophyll fluorescence SIF is calculated. Using a pre-defined physiologically constrained nonlinear transformation operator, a model was established to transform the fluorescence intensity (SIF) of pure chlorophyll into the maximum carboxylation rate (V). cmax25 Quantitative relationship; Based on the above, this embodiment limits the implementation method of the signal demixing and analog generation module; The module's workflow is divided into two core steps. The first step applies a pre-defined ecologically constrained mixed-pixel unmixing model to separate fluorescence signals. This model transcends traditional linear mixing in its design, incorporating ecological and physical corrections. It uses vegetation functional type coverage as the weighting basis for different vegetation components and introduces a pre-defined canopy radiative transfer correction operator and an ecological interaction term. The pre-defined canopy radiative transfer correction operator is a mathematical function used to correct for signal attenuation effects caused by shading and scattering during signal propagation in the vegetation canopy, based on established radiative transfer theories such as Beer-Lambert's law. The ecological interaction term is a function specifically designed to quantify the nonlinear signal superposition effects between different vegetation functional types due to resource competition or physical shading. By solving the model, the pure chlorophyll fluorescence (SIF) corresponding to each vegetation functional type can be calculated from the total fluorescence of the mixed pixels. The second step involves applying a pre-defined physiologically constrained nonlinear transformation operator to establish a transformation from pure chlorophyll fluorescence (SIF) to the maximum carboxylation rate (V). cmax25 The quantitative relationship between them; the preset physiological constraint nonlinear transformation operator refers to a mechanism based on the principles of photosynthetic physiology that establishes the relationship between pure chlorophyll fluorescence SIF and maximum electron transport rate (J). max25A mathematical model of the nonlinear relationship between fluorescence and GPP; the underlying mechanism is that fluorescence and GPP share the same electron transport chain, but their distribution ratio is dynamically regulated by environmental stress, so their relationship is not a simple linear proportional relationship; the operator is specifically used to describe this complex nonlinear transformation. By introducing an unmixing model constrained by ecological mechanisms and a transformation operator constrained by physiological constraints, this embodiment significantly improves the physical authenticity of the observation data processing; it can more accurately separate the fluorescence signals of different vegetation, thus providing higher quality and lower uncertainty observation input for the subsequent assimilation and inversion module, which is a key step in improving the final GPP inversion accuracy.

[0021] Example 3 Ecological interaction terms are modeled as functions with the coverage, leaf area index and exclusive competition coefficient between different vegetation functional types as inputs, and are used to describe the nonlinear effects caused by resource competition or shading between different vegetation functional types. The physiological constraint nonlinear transformation operator includes a nonlinear adjustment term, which is related to temperature and water vapor pressure difference in environmental stress data and is used to characterize the regulation of electron transport chain by environmental stress. In this embodiment, the composition of the ecological interaction term and the physiological constraint nonlinear transformation operator is further described in detail. These two features work together to improve the accuracy of the observation operator. To further clarify, the modeling method for ecological interactions is specified; ecological interactions are symbolically represented as follows: Its function is to correct the nonlinear fluorescence effect caused by the coexistence of different vegetation communities; in this embodiment, it is modeled as a function with multiple ecological variables as input, and can be expressed mathematically as follows: ; in, and They are the first in each pixel species and first The coverage of vegetation functions is a dimensionless parameter, which is derived from high-resolution land cover data acquired by the observation data acquisition module. and These are the leaf area indices of the two vegetation functional types, which are dimensionless parameters used to characterize the density of the canopy, and are derived from remote sensing inversion products or model simulations. This is a specific competition coefficient, a dimensionless parameter, used to quantify a particular competitive relationship between two vegetation types, such as the shading effect between C3 herbaceous plants and C4 crops. It is derived from ground quadrat experimental data or prior ecological knowledge. To ensure dimensional consistency, the function... A function containing empirical parameters in units of fluorescence, to ensure The output dimension is consistent with that of fluorescence; To further clarify, the function It can take a non-restrictive specific form, for example, modeling it as a linear summation of the pairwise interactions between different vegetation functional types, for any two different vegetation functional types within a pixel. and The interaction term can be expressed as: ; in, For any two different vegetation functional types i and j within a pixel, there is an interaction term; It is an empirical conversion factor with the dimension of fluorescence, used to convert dimensionless interaction intensities into fluorescence units. The factor can be obtained through model calibration. Therefore, the total ecological interaction term It can be the sum of all interaction terms between different vegetation functional type pairs: ; in, For the overall ecological interaction items; For interaction terms between different vegetation functional type pairs; Representing different vegetation functional types; the form describes the competitive effects between vegetation types, such as shading and canopy density, and the canopy density of each different vegetation functional type, in terms of coverage. Leaf area index The product of the products is approximately proportional to the product of the products, which is a feasible modeling method that conforms to ecological common sense; The function is designed to capture the nonlinear gain or loss of fluorescence signals caused by competition for resources such as light, water, and nutrients among vegetation or by physical shading between them. The core component of the physiological constraint nonlinear transformation operator is specified; the key to the operator lies in its inclusion of a nonlinear adjustment term; the nonlinear adjustment term is symbolically represented as... Its function is to characterize the dynamic regulation of the electron transport chain in photosynthesis by environmental stress, thereby affecting the energy distribution ratio between fluorescence and GPP; the regulation term is designed to correlate with the temperature in the environmental stress data acquired by the observation data acquisition module. and water vapor pressure difference Closely related; ; in, This is a nonlinear adjustment term, which characterizes the dynamic regulation of the electron transport chain in photosynthesis by environmental stress; the function The model is based on recognized plant physiological models. For example, excessively high or low temperatures, as well as large vapor pressure differences (i.e., dry air), can inhibit photosynthesis and alter its function. The value is used to adjust the conversion efficiency from fluorescence to GPP; As a power exponent, it is a dimensionless value, and the function... Dimensional input and Convert to dimensionless output; To give a non-restrictive example, a function It can be constructed as the product of two independent stress functions: ; in, : Nonlinear adjustment term; The product of two independent stress functions; Water vapor pressure stress function; Temperature stress function A parabolic curve, commonly used for photosynthesis, can be adopted: ; in, Temperature in environmental stress data; here This is the optimal temperature for the growth of functional vegetation. It is the lowest temperature at which it can perform photosynthesis; the parameter can be obtained from a physiological parameter database, and the water vapor pressure stress function. The effect of air drying on stomatal conductance can be characterized by an exponential decay form: ; in, : Water vapor pressure difference stress function; : Water vapor pressure difference in environmental stress data; It is a vegetation-specific water stress sensitivity coefficient, with dimensions of water vapor pressure difference. The reciprocal of the dimension, for example, hPa⁻¹, is used to ensure that the exponent term is dimensionless. In this way, The value can respond specifically and non-linearly to changes in the environment; The specific modeling of ecological interaction terms enables the quantification of nonlinear competition effects between different vegetation types during the signal unmixing process. This is particularly important in complex surface environments such as agroforestry-pastoral complexes, significantly improving the separation accuracy of pure chlorophyll fluorescence (SIF). Simultaneously, by introducing nonlinear adjustment terms directly related to temperature and humidity, the fluorescence-to-GPP conversion process can dynamically respond to real environmental stresses, overcoming the significant errors caused by using fixed conversion coefficients in traditional methods. The combination of these two approaches ensures the physical authenticity and accuracy of the GPP observations input to the assimilation system. In the processing of observed signals, this scheme demonstrates significant progress. Existing technologies typically employ linear spectral mixing models to process mixed remote sensing pixels. This method assumes that the total signal is a simple weighted sum of the component signals, neglecting the complex interactions between vegetation functional types. This scheme introduces a mixed pixel unmixing model based on ecological mechanism constraints, particularly defining the ecological interaction terms within it. This allows for a more precise description of this physical process; Coverage was modeled as a function of different vegetation types. Leaf area index and exclusive competition coefficient Related functions; for example, in a mixed pixel containing both tall trees and low-lying herbs, the trees have a light shading effect on the herbs, which is not a simple linear superposition of signals; the ecological interaction term of this scheme can quantify the nonlinear attenuation of herb fluorescence signals caused by species shading, thereby calculating a more realistic pure chlorophyll fluorescence SIF value, which is impossible to achieve with traditional linear models. This scheme, in the step of using pure chlorophyll fluorescence SIF to observe and constrain the model state, abandons the traditional approach of using fixed, linear relationships; its core lies in the application of a method that includes a nonlinear adjustment term. Physiologically constrained nonlinear transformation operators were used to establish a transformation from pure chlorophyll fluorescence (SIF) to the maximum carboxylation rate (V). cmax25 Quantitative relationship between them; adjustment term The practical significance lies in capturing the dynamic regulation of plant photosynthetic physiology by environmental stress. When plants are under drought or high temperature stress, the energy distribution of the photosynthetic electron transport chain tends to be heat dissipation rather than photochemical reaction, resulting in a change in the ratio of fluorescence to GPP. This scheme, through this nonlinear adjustment term, enables the model to accurately characterize the nonlinear relationship between SIF and photosynthesis under environmental stress, thereby allowing the assimilation system to infer a more realistic vegetation physiological state based on real SIF observations, thus improving the reliability of the inversion results under extreme climate events.

[0022] Example 4 The state component decoupling module is specifically used for: Based on the photosynthetically active radiation obtained by the observation data acquisition module, the fast component of the light response is determined. By applying a pre-defined temporal convolution model and integrating the historical environmental stress, the memory component is determined. This embodiment limits the implementation method of the state component decoupling module; The module's workflow is broken down into two parallel steps, each used to determine different components of the GPP; the first step is to determine the fast component of the photoresponse. The fast component of the light response is defined as the portion of the GPP that responds instantly to changes in illumination. In this embodiment, its determination is based on photosynthetically active radiation (PAR) data acquired by the observation data acquisition module. A standard light response curve model, such as the Mie equation or a non-rectangular hyperbola model, is used to directly calculate the instantaneous PAR from sub-hourly PAR data. The value; the component is not included in subsequent state evolution prediction because it is considered a fast, memoryless process; This module calculates the fast component based on photosynthetically active radiation. This module uses a temporal convolution model to calculate memory components. Slow component It is defined as a core state variable that needs to be predicted through model evolution, and its initial value can be set or obtained from prior knowledge; The second step is to determine the memory components. The memory component is defined as the part of GPP that reflects the cumulative effect of historical environmental stress. In this embodiment, it is determined by applying a preset temporal convolution model. The preset temporal convolution model refers to a mathematical tool used to describe the cumulative effect. The model obtains the current memory component value by performing an integral operation on the environmental stress sequence over a historical period. The mathematical expression is: ; in, It is in time The memory component has the same dimensions as GPP; It is a memory kernel function that defines how the influence of past stresses diminishes over time, with the dimension being GPP / time; In the past time The dimensionless environmental stress index integrates multiple environmental stress factors such as temperature and moisture, and is derived from environmental stress data acquired by the observation data acquisition module. :time; Over time, the integral variable; To achieve the dimensionless environmental stress index It can be defined as 1 minus the nonlinear adjustment term defined in the aforementioned embodiments. Because it quantifies the comprehensive impact of environmental stress on the allocation of photosynthetic electron transport chains, the calculation formula can be: ; in, In the past time The dimensionless environmental stress index; In the past time The nonlinear adjustment term; In the past time Temperature; In the past time Water vapor pressure difference; function The form of the stress index has been explained in the above embodiments, and this definition ensures that the stress index is... The value varies between 0 (no stress) and 1 (complete inhibition of physiological activity), and maintains an intrinsic consistency with the physiological regulatory mechanism of fluorescence to GPP conversion. The physical meaning of integration is to multiply the environmental stresses from all moments in the past to the present by a weight that decays over time, namely the memory kernel function, and sum them up to obtain the overall impact on the current photosynthetic capacity. Through this decoupling method, this embodiment clearly separates the distinctly different response processes in GPP; it directly calculates the rapidly changing light response components, while modeling and assimilating the slow and memory components as core state variables, which greatly simplifies the complexity of the state evolution model and allows the assimilation framework to focus on capturing and predicting the slow physiological processes with memory that are crucial to ecosystem stability, thereby improving the accuracy and computational efficiency of the model prediction. This scheme achieves a theoretical breakthrough in describing the response mechanisms of ecosystems to the environment; traditional models often neglect the memory effect of vegetation on environmental stress. This scheme decomposes the total GPP by constructing a state component decoupling module and decomposes the memory component within it. The calculation is performed using a temporal convolution model: ; The physical implication of the formula is that current photosynthetic capacity is the result of all historical environmental stresses. Through a memory kernel function The result of cumulative effect.

[0023] The temporal convolution model employs a double-exponential memory kernel function to capture the composite memory characteristics of rapid vegetation recovery and slow adaptation. This embodiment describes the memory kernel function used in the temporal convolution model. A design was carried out; To more realistically simulate the complex memory behavior of vegetation, the memory kernel function in this embodiment adopts a double-exponential form. A double-exponential memory kernel function is a function consisting of the linear superposition of two exponentially decaying terms. The design motivation is to capture the memory characteristics of two different time scales coexisting in the physiological response of vegetation. The mathematical form is: ; In the function: : Memory kernel function; and These are two weighting coefficients, with the dimension of GPP / time. Their relative magnitudes determine the relative importance of the two memory processes. The initial values ​​can be obtained from the literature on control experiments of specific vegetation functional types. and These are two characteristic time scales, with the dimension of time; among them, It is usually set to a small value, such as several days, to capture the rapid recovery characteristics of vegetation after the relief of short-term stress, such as the rapid rebound of photosynthetic capacity after rainfall following drought. This is then set to a relatively large value, such as several weeks to several months, to capture the physiological structural adjustments that vegetation makes in response to long-term environmental changes, i.e., the complex memory characteristics of slow adaptation, such as changes in root growth or stomatal density in response to prolonged drought; parameters During the assimilation process, these parameters will be used as optimization parameters and will be learned and adjusted by the system based on actual observation data. As time goes by; The base of the natural logarithm, a standard mathematical constant; Compared to single-exponential kernel functions, which can only describe a single decay process, the double-exponential form used in this embodiment can simultaneously characterize two key memory mechanisms: rapid recovery and slow adaptation of vegetation. This design makes the calculation of memory components more consistent with biological reality and can more accurately simulate the GPP dynamics of vegetation under complex scenarios such as rapid recovery after flash drought or slow adaptation during seasonal drought, significantly improving the model's performance under extreme or variable climate conditions. Crucially, the memory kernel function Designed in double exponential form: ; This is not merely a mathematical construct, but has profound physiological significance; it represents that vegetation possesses memory on at least two time scales: one composed of characteristic time... For example, a rapid recovery process dominated by several days, and a recovery process characterized by time... For example, slow adaptation processes that typically last from weeks to months; for instance, a forest experiencing a brief flash drought may see its Gross Productivity (GPP) recover rapidly within days, which corresponds to... However, if a drought persists for an entire season, the internal physiological adjustments, such as root biomass allocation, are a slow adaptation process, and their effects can last for months. This corresponds to... This precise characterization of complex memory properties enables the model to capture the true dynamics of ecosystems under complex stress sequences, significantly outperforming existing single-timescale models.

[0024] The critical transition model of ecosystem state is a nonlinear stochastic differential equation based on bifurcation theory, used to describe the nonlinear response of ecosystems under environmental stress. This embodiment defines the form of the critical transition model for ecosystem states. The purpose of the ecosystem state critical transition model is to describe the dynamic evolution of the slow and memory components of the Global Productivity Process (GPP), especially the nonlinear behavior of the ecosystem as it approaches its carrying capacity limit or critical point. In this embodiment, the model is a nonlinear stochastic differential equation based on bifurcation theory. Bifurcation theory is a mathematical theory that studies the qualitative changes in the behavior of a system when its parameters change. Introducing it into the ecological model can effectively describe the sudden transition of the system. The core of the equation is derived from the traditional logistic growth model, and its form is: ; To ensure dimensional consistency in the formula, the original logistic growth term was modified to... In the equation, each term is defined as follows: It is the fast component of the photoresponse. It is a slow component. It is the sum of the two memory components. The state variables and rates of change of the system constitute the system. The dimension of is mass / (area·time²); It is the total GPP, which is equal to the sum of the fast, slow, and memory components; It is the intrinsic growth rate of the ecosystem, with the dimension 1 / time; It is the maximum GPP under environmental carrying capacity, and its dimensions are the same as GPP; It is a comprehensive, dimensionless environmental stress index, calculated from environmental stress data acquired by the observation data acquisition module; It is the critical stress threshold that the ecosystem can withstand. It is dimensionless and is a key parameter to be optimized in the assimilation and inversion process. It is the Herveside step function, when Its value is 1 if it is true, and 0 otherwise. It is a bifurcation parameter, dimensionless, representing the rate at which the control system collapses after crossing the critical point; It is an additive random noise term, representing random disturbances that the model fails to describe, and its dimensions are the same as those on the left side of the equation; To make the equation computable, on the and To provide a non-restrictive description, the comprehensive environmental stress index The stress index used in the memory component can be adopted. The same definition, namely: ,in, Comprehensive Environmental Stress Index; The stress index used in the memory component; to ensure that the quantification method of environmental stress is consistent throughout the model, additive random noise term. It is typically modeled as a Gaussian white noise process, with a discrete time step of... Within the context, the contribution of the additive random noise term to the state variable can be expressed as: ,in η is a random variable that follows a standard normal distribution with a mean of 0 and a variance of 1, while σ is the parameter to be optimized, representing the noise intensity. Its dimension is the same as η(t), which is GPP / time. The equation describes the state variables Evolution; under normal conditions The second term in the parentheses on the right side of the equation is 1, and the model exhibits a logistic growth pattern; under environmental stress... Exceeding the critical threshold hour, The function is activated, and the growth term of the equation is drastically weakened or even becomes negative, thus simulating a sudden decline or collapse in ecosystem productivity; By introducing nonlinear stochastic differential equations based on bifurcation theory, this system no longer merely fits the smooth changes in GPP, but possesses the ability to predict nonlinear abrupt changes in the ecosystem, i.e., critical transitions; this is achieved through continuous parameter optimization during the assimilation process. and and tracking state variables The model can provide early warnings of the risk of ecosystems nearing collapse due to increased environmental stress, providing crucial early warning information for the management and protection of ecosystems. The most significant advancement lies in its ability to predict the nonlinear behavior of ecosystems. Existing models are mostly linear or approximately linear, unable to predict critical transitions, i.e., abrupt changes in state, that may occur in ecosystems under sustained stress. This approach employs an ecosystem state critical transition model based on bifurcation theory as the core of its state evolution prediction module. ; The derivation of the equation is based on a modification of the standard logistic growth model to ensure dimensional consistency and completeness of physical meaning; among which, the terms The introduction of Herveside function is key; This constitutes a nonlinear switch in the system, the physical meaning of which is: when the comprehensive environmental stress index... Below the critical threshold When the stress exceeds the critical point, the term is 1, and the system evolves normally; once the stress exceeds the critical point, the term changes. This has a drastic inhibitory effect on the growth of the system, thus simulating the collapse of ecosystem productivity; the critical parameters are continuously optimized in the adaptive assimilation and inversion module. and This solution can not only reverse the current GPP, but also provide early warning of ecosystem stability, which is a forward-looking function that existing technologies do not have. Growth restrictions here Use total GPP The ecological hypothesis is that ecosystems can be used to support slow physiological processes. The increasing resources are subject to instantaneous competition from the current total photosynthetic output, namely the fast light response component. It also consumes system resources, thus limiting the growth of slow components. This setting allows the model to more realistically reflect the resource allocation and competition relationships between processes at different time scales within the system.

[0025] Example 5 A regional GPP assimilation and inversion system integrating ground-based and multi-source satellite chlorophyll fluorescence, characterized in that the adaptive assimilation and inversion module is specifically used for: Construct an assimilation framework with slow components and memory components as state variables; The state transition equations of the assimilation framework are based on the state evolution prediction module. The clean SIF observations provided by the signal demixing and simulation generation module are used as the observation inputs of the assimilation framework, and the state variables and model parameters are optimized through data assimilation. This embodiment provides a detailed explanation of the working mechanism of the adaptive assimilation and inversion module; The module's workflow is organized into a standard data assimilation framework, comprising three core steps. The first step involves constructing an assimilation framework with slow and memory components as state variables. This means that in the assimilation system, the core element to be estimated and predicted is the intrinsic physiological state of the vegetation, i.e. and It is not directly the total GPP; the assimilation framework can be used in algorithms such as sequence importance resampling particle filtering, ensemble Kalman filtering or variational assimilation. The second step uses the state evolution prediction module as the state transition equation of the assimilation framework; this means that during the prediction phase of the assimilation algorithm, the system will invoke the aforementioned nonlinear stochastic differential equation; the equation uses the current state variables ( Using as input, the predicted state for the next time step is calculated by forward integration. ) and its uncertainties; the steps constitute the model-driven physical evolution part of the assimilation process; The third step involves using the clean SIF observations as the input to the assimilation framework and optimizing the state variables and model parameters through data assimilation. During the update or analysis phase of the assimilation algorithm, the system introduces clean SIF observations provided by the signal demixing and simulation generation module. The assimilation algorithm compares the differences between the simulated SIF values ​​(after transformation by the observation operator) and the actual SIF observations, considering their uncertainties, to calculate the impact on the state variables. The optimal estimate; at the same time, the difference can also be used to optimize key parameters in the online optimization model, such as These features enable the model to better learn the behavior of ecosystems in specific regions. By constructing the assimilation module in this way, this embodiment dynamically integrates mechanistic model predictions with remote sensing data observations. Instead of simply fitting the data with a model, it allows the data to continuously constrain and correct the model's trajectory in a cyclical iterative process. This makes the GPP inversion results both conform to the basic laws of ecology, guaranteed by the state transition equation, and consistent with real-time satellite observations, thus obtaining GPP estimation results that are far more accurate and robust than single model simulations or simple empirical statistics.

[0026] The adaptive assimilation and inversion module is also used for: Based on the dominant vegetation functional type within the pixel, the corresponding parameters are retrieved from the preset physiological parameter library; Based on the retrieved parameters, the constraint terms in the assimilation cost function are dynamically adjusted to ensure that the model evolution conforms to the physiological laws of specific vegetation functional types. This embodiment enhances the adaptive characteristics of the adaptive assimilation and inversion module; The module includes two additional adaptive adjustment steps during assimilation and inversion. The first step retrieves corresponding parameters from a pre-defined physiological parameter library based on the dominant vegetation functional type within the pixel. This library is a pre-built database storing typical physiological parameters for different vegetation functional types (PFTs), such as C3 herbaceous, C4 crop, evergreen coniferous forest, and deciduous broadleaf forest. Parameters include, but are not limited to, intrinsic growth rate. Environmental carrying capacity Memory kernel parameters When processing each pixel, the system first identifies the dominant PFT based on high-resolution land cover data, and then selects the most suitable set of initial parameters for the pixel from the library. The adaptive module in this scheme uses a set of high-order ecological strategy parameters (such as intrinsic growth rate k, environmental carrying capacity C, etc.) to characterize the overall adaptation strategies of different vegetation functional types. These parameters are not traditional biophysical parameters, but rather specific parameters (such as V) required for dynamically generating and optimizing ecological process models (such as CLM) through internally preset mapping relationships. cmax25The initial values ​​and ranges of (Ω, etc.) are determined to achieve adaptive assimilation and inversion based on PFT and real-time stress state.

[0027] As a non-limiting example, entries in a physiological parameter library might contain the following: For C3 herbs, the parameter set might be: ; For evergreen coniferous forests, the parameter set might be: ; The initial values ​​will be further optimized during the assimilation process; The second step involves dynamically adjusting the constraint terms in the assimilation cost function based on the retrieved parameters to ensure that the model evolution conforms to the physiological laws of specific vegetation functional types. The assimilation cost function is the objective function that the assimilation algorithm aims to minimize when seeking the optimal solution. It typically includes a difference term between model predictions and observations, as well as a background error term. The innovation of this embodiment lies in introducing additional PFT-related constraint terms into the function. For example, for pixels dominated by C3 plants, the system strengthens the constraint weights related to the temperature response curve in the cost function because the photosynthesis of C3 plants is more sensitive to temperature changes. For pixels dominated by C4 plants, the constraint on the light saturation effect is strengthened because C4 plants usually have a higher light saturation point. For CAM plants, their unique diurnal acidity variation rhythm can be introduced as a strong temporal constraint. This strategy introduces ecological prior knowledge into the optimization process in the form of mathematical constraints. This dynamic adjustment can be achieved by modifying the assimilation cost function. To achieve this, a typical variational assimilation cost function takes the form of: ; in, : Assimilation cost function; Superscript T: matrix transpose; Superscript -1: matrix inverse; : Observation operator; here It is a state variable, such as ; It's a background scene. It is the observed value, namely pure chlorophyll fluorescence. , and These are the error covariance matrices of the background and observations, respectively. It is an observation operator, whose function is to measure the state variables of the model. (Right now This is converted into a quantity comparable to the observed value, namely simulated chlorophyll fluorescence. The innovation of this invention lies in the introduction and adjustment of constraints related to PFT. For example, constraints can be designed as follows: ; in, It is a function used to express specific physiological laws, and It is a weighting coefficient retrieved from the parameter library based on the dominant PFT type. For example, when processing C4 crop pixels, it can be retrieved from the parameter library for... Choose a larger value and let This indicates that the model state has not reached the light saturation level under strong light, thus strongly constraining the model to conform to the high light efficiency characteristics of C4 plants during optimization; : Weighting coefficients retrieved from the parameter library based on the dominant PFT type; A function used to express a specific physiological law; : State variables; Through this two-layer adaptive adjustment mechanism, the assimilation system achieves a high degree of intelligence and specificity; instead of using the same model and parameters to process all types of land surfaces in a one-size-fits-all manner, it configures an assimilation scheme that best matches the main physiological characteristics of its vegetation for each pixel; this deep integration of macro-ecological knowledge, PFT classification and micro-mathematical model, and cost function constraints greatly improves the applicability and accuracy of the model on heterogeneous land surfaces, making the final inverted regional GPP product more realistic and reliable in terms of the transition between different ecosystems and the internal details; The adaptive assimilation and inversion module of this scheme solves the problem of the one-size-fits-all approach of traditional models by introducing a parameterization scheme based on vegetation functional types. The system can retrieve specific parameters from a preset physiological parameter library according to the dominant vegetation functional type in a pixel, such as C3 and C4 plants, and dynamically adjust the constraint terms in the assimilation cost function. This means that the behavior of the model itself will be adaptively adjusted according to the physiological characteristics of its simulated objects. For example, it strengthens the constraint of adaptability to high light intensity for C4 crops and enhances the constraint of sensitivity to temperature for C3 forests. This strategy of embedding macro-ecological knowledge into the assimilation process in a mathematical form ensures the seamless transition of inversion results between different ecosystem types and the authenticity of internal details. Its accuracy and reliability are far superior to traditional models that use general parameters.

[0028] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A regional GPP assimilation and inversion system integrating ground-based and multi-source satellite chlorophyll fluorescence, characterized in that, include: The observation data acquisition module is used to acquire data on solar-induced chlorophyll fluorescence, photosynthetically active radiation, high-resolution land cover, and environmental stress from ground-based and remote sensing observations. The signal demixing and simulation generation module is used to demix the total fluorescence of mixed pixels into pure chlorophyll fluorescence (SIF) based on the vegetation functional coverage determined by high-resolution land cover data, and drive the ecological model to generate SIF and GPP simulation values ​​based on environmental stress data. The state component decoupling module is used to decouple the total GPP into fast optical response components, slow optical response components, and memory components. The state evolution prediction module is used to predict the dynamic evolution of the slow component and the memory component based on a preset ecosystem state critical transition model. The adaptive assimilation and inversion module is used to invert the regional GPP by using the slow component and memory component as state variables, the SIF observations as observation data, and combining the prediction results of the state evolution prediction module.

2. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 1, characterized in that, The signal demixing and analog generation module is specifically used for: By applying a pre-defined ecological mechanism-constrained mixed pixel unmixing model, combined with vegetation functional coverage, a pre-defined canopy radiative transfer correction operator, and ecological interaction terms, the pure chlorophyll fluorescence SIF is calculated. Using a pre-defined physiologically constrained nonlinear transformation operator, a model was established to transform the fluorescence intensity (SIF) of pure chlorophyll into the maximum carboxylation rate (V). cmax25 The quantitative relationship.

3. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 2, characterized in that, The ecological interaction term is modeled as a function with the coverage, leaf area index and exclusive competition coefficient between different vegetation functional types as inputs, and is used to describe the nonlinear effects caused by resource competition or shading between different vegetation functional types.

4. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 2, characterized in that, The physiological constraint nonlinear transformation operator includes a nonlinear adjustment term, which is related to temperature and water vapor pressure difference in the environmental stress data and is used to characterize the regulation of the electron transport chain by environmental stress.

5. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 1, characterized in that, The state component decoupling module is specifically used for: Based on the photosynthetically active radiation obtained by the observation data acquisition module, the fast component of the light response is determined. By applying a pre-defined temporal convolution model and integrating the historical environmental stress, the memory component is determined.

6. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 5, characterized in that, The temporal convolution model employs a double-exponential memory kernel function to capture the composite memory characteristics of rapid vegetation recovery and slow adaptation.

7. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 1, characterized in that, The critical transition model of the ecosystem state is a nonlinear stochastic differential equation based on bifurcation theory, used to describe the nonlinear response of the ecosystem under environmental stress.

8. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 1, characterized in that, The adaptive assimilation and inversion module is specifically used for: Construct an assimilation framework with slow components and memory components as state variables; The state transition equations of the assimilation framework are based on the state evolution prediction module. The clean SIF observations provided by the signal demixing and simulation generation module are used as the observation inputs to the assimilation framework, and the state variables and model parameters are optimized through data assimilation.

9. The regional GPP assimilation and inversion system for fusion of ground-based and multi-source satellite chlorophyll fluorescence according to claim 8, characterized in that, The adaptive assimilation and inversion module is also used for: Based on the dominant vegetation functional type within the pixel, the corresponding parameters are retrieved from the preset physiological parameter library; Based on the retrieved parameters, the constraint terms in the assimilation cost function are dynamically adjusted to ensure that the model evolution conforms to the physiological laws of specific vegetation functional types.

Images