Method and system for estimating mesophyll conductivity of plant by utilizing sunlight-induced chlorophyll fluorescence

Abstract

The invention discloses a method and a system for estimating the mesophyll conductivity of a plant by utilizing sunlight-induced chlorophyll fluorescence, and relates to the technical field of plant physiological prediction. The method comprises the following steps: obtaining a plant mesophyll conductance prediction factor, taking the prediction factor as input, and outputting the plant mesophyll conductance based on a random forest model. Wherein the plant mesophyll conductivity predictive factor comprises a photosystem II reaction center opening ratio, a photosystem II maximum photochemical quantum efficiency, an atmospheric carbon dioxide concentration, a leaf temperature, a saturation vapor pressure difference and photosynthetically active radiation. According to the method, the qL is estimated by adopting the OC redox model, so that direct measurement of the qL in a complex experiment is avoided, and the application feasibility of the qL in global-scale ecological research is remarkably improved.

Description

Technical Field

[0001] The present invention relates to a method and system for estimating leaf mesophyll conductance using solar-induced chlorophyll fluorescence, and belongs to the technical field of plant physiological prediction. Background Art

[0002] During photosynthesis, carbon dioxide first diffuses from the surrounding atmosphere into the substomatal cavity through stomata, and then diffuses through the mesophyll layer to the carboxylation site within the chloroplast.

[0003] For a long time at different spatio-temporal scales, stomatal conductance (gs), as an indicator characterizing the degree of stomatal opening, has always been the focus of attention of physiologists, ecologists, and modelers. This is because stomatal conductance not only controls the exchange of water and carbon dioxide between plants and the atmosphere but also profoundly affects the response of plants to environmental factors. In contrast, mesophyll conductance (gm) reflects the diffusion ability of carbon dioxide within mesophyll cells and is considered equally important as stomatal conductance in limiting photosynthesis. However, the research on mesophyll conductance has mainly been limited to the leaf scale.

[0004] At the leaf scale, the measurement of mesophyll conductance usually relies on techniques such as the combination of gas exchange and chlorophyll fluorescence, stable carbon isotope discrimination method, curve fitting method, or leaf anatomical structure modeling. These methods provide important means for revealing the mechanism of carbon dioxide diffusion within leaves, evaluating the response of mesophyll conductance to environmental changes, and quantifying the impact of anatomical structure. However, extending these methods from the leaf scale to the canopy scale and even larger scales still faces many challenges.

[0005] Currently, most large-scale carbon cycle models only explicitly consider stomatal conductance and assume that mesophyll conductance is infinite. This assumption may lead to an overestimation of the carbon dioxide concentration (Cc) at the carboxylation site of chloroplasts, and thus an underestimation of the carbon dioxide fertilization effect in the Earth system.

[0006] Therefore, it is crucial to develop a method for estimating mesophyll conductance applicable to all scales.

[0007] With the rapid development of chlorophyll fluorescence technology, photosynthetic physiologists have gradually shifted their research focus to the light reaction process of photosynthesis. This shift is mainly due to the significant expansion potential of light reaction parameters (especially solar-induced chlorophyll fluorescence - SIF) at spatio-temporal scales. As a light signal closely related to photosynthesis, SIF can be remotely obtained through high-resolution spectral sensors and advanced algorithms, and its emitted light and reflected light can be effectively separated. This technological progress makes it possible to quickly collect SIF data from satellites, ground observation towers, and airborne platforms, thus realizing multi-dimensional measurements from the leaf to the canopy, regional, and even global scales.

[0008] However, when estimating photosynthesis using a light - reaction - based model, it is still necessary to accurately input the carbon dioxide concentration at the carboxylation sites of chloroplasts. For example, in the light - reaction mechanism model of photosynthesis (MLR - SIF model), using the intercellular carbon dioxide concentration instead of the carbon dioxide concentration at the carboxylation sites of chloroplasts may lead to an underestimation of photosynthesis. Therefore, in order to accurately estimate photosynthesis through light - reaction parameters and promote the wide application of the MLR - SIF model at large scales, it is necessary to pre - estimate the carbon dioxide concentration at the carboxylation sites of chloroplasts from the perspective of light reactions using a mesophyll conductance model.

[0009] Theoretically, the light reaction and carbon reaction of photosynthesis are in a dynamic equilibrium state. During the light reaction, the production of reducing energy (ATP / NADPH) is determined by the linear electron transfer rate, while its consumption is affected by the photosynthetic capacity and carbon dioxide diffusion during the carbon reaction. Research has shown that the open ratio of photosystem II reaction centers ( ), which reflects the redox state of chloroplast quinone A (QA), is the rate - limiting step of electron transfer and can regulate the balance between the light reaction and carbon reaction of photosynthesis. In addition, the photosynthetic capacity parameter is closely related to , and gs can also be modeled through . Stomatal conductance, mesophyll conductance, and maximum carboxylation rate coordinate with each other in determining photosynthesis.

[0010] Currently, is usually derived by directly measuring chlorophyll fluorescence parameters through saturation pulse amplitude - modulation (PAM) fluorometry. However, the applicable range of this method is mainly limited to the leaf scale. In studies at scales above the leaf level, scientists mainly come from fields such as ecology, the earth biosphere, or carbon cycling. They usually do not focus on the research of physiological parameters such as , and also lack the professional skills to perform measurements using PAM fluorometry.

[0011] Therefore, developing a method that can obtain without relying on complex experimental measurements will help improve the application feasibility of in ecological research at the population, regional, and even global scales. Summary of the Invention

[0012] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method and system for estimating the mesophyll conductance of plants using solar - induced chlorophyll fluorescence, which does not rely on complex experiments to directly measure and obtain , but uses the OC redox model for estimation, improving the application feasibility of in ecological research at the global scale.

[0013] To achieve the above object, the present invention is implemented by the following technical solutions:

[0014] On the one hand, the present invention provides a method for estimating leaf mesophyll conductance of plants using solar-induced chlorophyll fluorescence, including:

[0015] Obtaining the prediction factors of plant leaf mesophyll conductance;

[0016] Using the prediction factors of plant leaf mesophyll conductance as input and outputting the plant leaf mesophyll conductance based on a random forest model;

[0017] Among them, the prediction factors of plant leaf mesophyll conductance include the open ratio of photosystem II reaction centers, the maximum photochemical quantum efficiency of photosystem II, atmospheric carbon dioxide concentration, leaf temperature, saturation vapor pressure deficit, and photosynthetically active radiation.

[0018] Furthermore, the maximum photochemical quantum efficiency of photosystem II is set as a constant;

[0019] The atmospheric carbon dioxide concentration, leaf temperature, vapor pressure deficit, and photosynthetically active radiation are obtained through measurement;

[0020] The open ratio of photosystem II reaction centers is calculated through the relationship function between solar-induced chlorophyll fluorescence, linear electron transfer rate, and the open ratio of photosystem II reaction centers, where the solar-induced chlorophyll fluorescence is obtained through measurement.

[0021] Furthermore, the expression of the relationship function between the linear electron transfer rate and solar-induced chlorophyll fluorescence is:

[0022] ;

[0023] Among them, represents the linear electron transfer rate, represents the maximum photochemical quantum efficiency of photosystem II, , where represents the intrinsic thermal dissipation constant, represents the fluorescence rate constant, represents the open ratio of photosystem II reaction centers, represents solar-induced chlorophyll fluorescence.

[0024] Furthermore, the relationship function between the linear electron transfer rate and the open ratio of photosystem II reaction centers is obtained by fitting the open-close redox model of photosynthetic electron transfer rate using the light response curve and the CO2 response curve, and its expression is:

[0025] ;

[0026] ;

[0027] ;

[0028] ;

[0029] Among them, represents the linear electron transfer rate, U represents the maximum oxidation potential of the cytochrome b6f complex for the plastoquinone and plastohydroquinone pools, represents the standardized temperature response function described by the redox reaction rate, represents the S-shaped function of the photosynthetically active radiation absorbed by the leaf, represents the redox balance function between photosystem II and cytochrome, represents the proportion of the reversible photosystem II reaction center, represents the open ratio of the photosystem II reaction center, represents the primary resistance of photosystem II electron transfer, represents the secondary resistance of photosystem II electron transfer, represents the standard reference temperature, represents the leaf temperature, represents the comprehensive temperature sensitivity parameter related to the Gibbs free activation energy, represents the stoichiometric parameter, , represents the empirical parameter.

[0030] Furthermore, the unknown relationship parameters in the relationship function between the linear electron transfer rate and the open ratio of the photosystem II reaction center include the maximum oxidation potential of the cytochrome b6f complex for the plastoquinone and plastohydroquinone pools, the proportion of the reversible photosystem II reaction center, the primary resistance of photosystem II electron transfer, the secondary resistance of photosystem II electron transfer, the comprehensive temperature sensitivity parameter related to the Gibbs free activation energy, the stoichiometric parameter, and the empirical parameter;

[0031] The determination method of the unknown relationship parameters includes:

[0032] Obtain multiple sets of corresponding true values of the linear electron transfer rate and the true values of the open ratio of the reaction centers in photosystem II;

[0033] Successively use multiple sets of true values of the open ratio of the photosystem II reaction centers as inputs, and predict the linear electron transfer rate through the relationship function between the linear electron transfer rate and the open ratio of the photosystem II reaction center combined with an optimization algorithm to obtain the predicted value of the linear electron transfer rate;

[0034] Calculate the objective function based on the predicted value and the true value of the linear electron transfer rate, and the value of the unknown relationship parameter corresponding to the minimum objective function value is the optimal solution.

[0035] Further, the optimization algorithm includes one or more of differential evolution optimization algorithm, basin hopping optimization algorithm, global optimization algorithm, dual annealing optimization algorithm, and DIRECT optimization algorithm.

[0036] Further, the objective function is the sum of squared errors between the predicted value and the true value of the linear electron transfer rate.

[0037] Further, the method for determining the plant mesophyll conductance prediction factor includes:

[0038] The relationship factors of mesophyll conductance obtained from the variable J method model include net photosynthetic rate, linear electron transfer rate, leaf temperature, and intercellular carbon dioxide concentration;

[0039] The relationship factor of net photosynthetic rate obtained from the photosynthesis physiological and biochemical model is the maximum carboxylation efficiency, and the maximum carboxylation efficiency is estimated from the proportion of open reaction centers in photosystem II and solar-induced chlorophyll fluorescence;

[0040] The relationship factors of linear electron transfer rate obtained from the MLR-SIF model are the proportion of open reaction centers in photosystem II, solar-induced chlorophyll fluorescence emitted by photosystem II, and the maximum photochemical quantum efficiency of photosystem II. Among them, the solar-induced chlorophyll fluorescence emitted by photosystem II is a function of photosynthetically active radiation;

[0041] The relationship factors of intercellular carbon dioxide concentration obtained from Fick's first law are atmospheric carbon dioxide concentration, net photosynthetic rate, and stomatal conductance. Among them, stomatal conductance is modeled by the proportion of open reaction centers in photosystem II, atmospheric carbon dioxide concentration, and saturation vapor pressure deficit;

[0042] Combining all the relationship factors is the plant mesophyll conductance prediction factor.

[0043] Further, it also includes pre-training the random forest model, and the pre-training method includes:

[0044] S1. Obtain a data set, which consists of multiple groups of corresponding plant mesophyll conductance prediction factors at the leaf scale and plant mesophyll conductance at the leaf scale;

[0045] S2. Divide the data set into a training set and a test set;

[0046] S3. Use the training set data as input to train the random forest model, and adjust the model parameters according to the loss function during training to obtain a preliminary random forest model;

[0047] S4. Use the test set data as input, test the random forest model to obtain test results, calculate evaluation metrics based on the test results. If the evaluation metrics are worse than the set threshold, repeat S3 - S4 until the evaluation metrics are better than the set threshold to obtain a pre-trained random forest model.

[0048] On the other hand, the present invention also provides a system for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence, including:

[0049] A memory for storing instructions;

[0050] A processor for executing the instructions, enabling the system to execute the method for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence as described in any one of the above.

[0051] Compared with the prior art, the beneficial effects achieved by the present invention:

[0052] The present invention determines the predictors of plant mesophyll conductance by combining multiple models and uses a pre-trained random forest model to quickly estimate mesophyll conductance, achieving seamless expansion from leaf to canopy, regional, and even global scales. This method significantly improves the estimation accuracy of global-scale photosynthetic carbon assimilation and provides more reliable data support for carbon cycle research.

[0053] Traditional carbon cycle models usually assume that mesophyll conductance is infinite, ignoring its limiting effect on photosynthesis. By introducing an accurate estimation of mesophyll conductance, the present invention can more accurately quantify the carbon dioxide concentration at the carboxylation sites of chloroplasts, thereby reducing the underestimation of the carbon dioxide fertilization effect and providing an important basis for improving carbon cycle models.

[0054] Based on solar-induced chlorophyll fluorescence (SIF) data, large-scale and high-efficiency data collection is achieved by combining remote sensing technology. Compared with the traditional saturation pulse amplitude modulation (PAM) fluorometry that relies on complex experiments, the present invention does not require professional personnel for on-site measurement, significantly reducing the cost and difficulty of data acquisition, while expanding the data coverage.

[0055] By using the OC redox model to estimate the open ratio q of photosystem II reaction centers L , direct measurement of complex experiments is avoided, simplifying the data acquisition process. This innovation enables researchers without professional skills in PAM fluorometry to easily conduct global-scale ecological research, greatly promoting the research efficiency and application breadth in the field of physiological ecology.

[0056] Through a pre-trained random forest model, the mesophyll conductance of plants can be estimated quickly and accurately, which is applicable to a variety of research scenarios from the leaf to the global scale. This method not only improves the estimation efficiency but also provides strong tool support for plant physiological ecology and carbon cycle research.

[0057] Combining plant physiology, ecology, remote sensing technology, and machine learning provides new ideas and methods for interdisciplinary research. By integrating multi-source data and multiple models, the present invention opens up new avenues for global-scale ecological research and promotes the collaborative development of related fields. Detailed implementation manners

[0058] The present invention is further described below. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and cannot be used to limit the protection scope of the present invention.

[0059] Embodiment 1:

[0060] An embodiment of the present invention provides a method for estimating the mesophyll conductance of plants using solar-induced chlorophyll fluorescence, including the following steps:

[0061] Obtain the prediction factors of plant mesophyll conductance. In this embodiment, the determination of the prediction factors of plant mesophyll conductance is mainly obtained by integrating multiple models. Specifically:

[0062] According to the variable J method model, the mesophyll conductance gm is the net photosynthetic rate , linear electron transport rate , leaf temperature (the main factor affecting the carbon dioxide compensation point), and the intercellular carbon dioxide concentration of the function, and further analysis:

[0063] According to the photosynthesis physiological and biochemical model (FvCB), the net photosynthetic rate is mainly affected by the maximum carboxylation efficiency , and the maximum carboxylation efficiency is linearly correlated with the maximum electron transport rate , the open ratio of the photosystem II reaction center and the solar-induced chlorophyll fluorescence SIF can be used to estimate the maximum carboxylation efficiency .

[0064] According to the MLR-SIF model, the linear electron transport rate can be calculated by the open ratio of the photosystem II reaction center , the SIF emitted by PSII, and the maximum photochemical quantum efficiency of photosystem II , where the SIF emitted by PSII is a function of the photosynthetically active radiation PAR.

[0065] According to Fick's first law, the intercellular carbon dioxide concentration can be calculated by the formula C i = C a – A n / g s where the stomatal conductance g s can be modeled by the open ratio of photosystem II reaction centers , the atmospheric carbon dioxide concentration and the vapor pressure deficit VPD.

[0066] Based on the above relationships, the predictors of the mesophyll conductance of the plants in this embodiment include the open ratio of photosystem II reaction centers , the maximum photochemical quantum efficiency of photosystem II , the atmospheric carbon dioxide concentration , the leaf temperature , the vapor pressure deficit VPD and the photosynthetically active radiation PAR.

[0067] At the canopy scale, among all the predictors of the mesophyll conductance of plants, the maximum photochemical quantum efficiency of photosystem II is one of the key parameters of the light reaction, which directly affects the electron transport rate required to support the biochemical reactions of photosynthesis. In the biochemical photosynthesis model, it is usually set as a constant 0.83, the atmospheric carbon dioxide concentration can be set as 410 µmol / mol, the leaf temperature can be approximately replaced by the air temperature, the vapor pressure deficit VPD can be measured using a commercial instrument (such as HMP-35, Vaisala, Helsinki, Finland), and the photosynthetically active radiation PAR can be measured using a photosynthetically active radiation quantum meter. Only the open ratio of photosystem II reaction centers cannot be directly measured.

[0068] To extend the mesophyll conductance estimation method based on the random forest model to the canopy and above scales, it is necessary to estimate the open ratio of photosystem II reaction centers

[0069] In this embodiment, the open ratio of photosystem II reaction centers is calculated by combining the solar-induced chlorophyll fluorescence with the relationship function between the linear electron transport rate and the open ratio of photosystem II reaction centers , and the relationship function between the linear electron transport rate and the solar-induced chlorophyll fluorescence SIF, where the solar-induced chlorophyll fluorescence is obtained by measurement.​

[0070] The relationship between the linear electron transfer rate and sunlight-induced chlorophyll fluorescence is expressed as follows:

[0071] ;

[0072] in, represents the linear electron transfer rate, represents the maximum photochemical quantum efficiency of photosystem II, ,in, represents the intrinsic heat dissipation constant, represents the fluorescence rate constant, Indicates the open ratio of the photosystem II reaction center, represents sunlight-induced chlorophyll fluorescence.

[0073] The parameters involved in the relationship function between the linear electron transfer rate and sunlight-induced chlorophyll fluorescence can all be directly measured or obtained through conventional settings.

[0074] sunlight-induced chlorophyll fluorescence coupled with linear electron transport rate Ratio of photosystem II reaction center opening The relationship function is obtained by fitting the open-close redox model of photosynthetic electron transfer rate using the light response curve and the CO2 response curve, and its expression is:

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] in, represents the linear electron transfer rate, U represents the maximum oxidation potential of the cytochrome b6f complex to the plastoquinone and plastohydroquinone pools, represents the normalized temperature response function describing the redox reaction rate, is the sigmoid function representing the photosynthetically active radiation absorbed by the leaves, represents the redox balance function between photosystem II and cytochrome, represents the proportion of reversible photosystem II reaction centers, represents the proportion of open reaction centers in photosystem II, represents the primary resistance of photosystem II electron transport, represents the secondary resistance of photosystem II electron transport, Indicates the standard reference temperature, represents the leaf temperature, represents the comprehensive temperature sensitivity parameter related to the Gibbs free activation energy, represents the stoichiometric parameter, 、 represents the empirical parameter.

[0080] sunlight-induced chlorophyll fluorescence coupled with linear electron transport rate Ratio of photosystem II reaction center opening The unknown relationship parameters in the relationship function include the maximum oxidation potential of the cytochrome b6f complex on plastoquinone and plastohydroquinone pools, the proportion of reversible photosystem II reaction centers, the primary resistance of photosystem II electron transfer, the secondary resistance of photosystem II electron transfer, the comprehensive temperature sensitivity parameter related to Gibbs free activation energy, the stoichiometric parameter, and the empirical parameter.

[0081] The values of unknown parameters can be determined by the following methods, including:

[0082] Multiple sets of corresponding true values of linear electron transfer rates and true values of photosystem II reaction center open ratios are obtained, which can be retrieved from an existing database in this embodiment.

[0083] Multiple sets of true values of the open ratio of the photosystem II reaction center are sequentially used as inputs, and the linear electron transfer rate is predicted by combining the relationship function between the linear electron transfer rate and the open ratio of the photosystem II reaction center with an optimization algorithm to obtain a predicted value of the linear electron transfer rate. The optimization algorithm can select one or more of the differential evolution optimization algorithm, basin jumping optimization algorithm, global optimization algorithm, double annealing optimization algorithm and DIRECT optimization algorithm.

[0084] The objective function is calculated based on the predicted linear electron transfer rate and the true linear electron transfer rate. In this embodiment, the objective function is the sum of squares of the errors between the predicted linear electron transfer rate and the true linear electron transfer rate.

[0085] The value of the unknown relationship parameter corresponding to the minimum objective function value is the optimal solution, which is the final determined value.

[0086] After the unknown relationship parameters are determined, the linear electron transfer rate Ratio of photosystem II reaction center opening relationship function, linear electron transfer rate The open ratio of the photosystem II reaction center can be calculated by linking it with the relationship function of sunlight-induced chlorophyll fluorescence SIF. .

[0087] Taking all plant mesophyll conductance predictors as inputs, plant mesophyll conductance is obtained based on the output of a random forest model.

[0088] This embodiment further includes pre-training the random forest model, and the pre-training process includes:

[0089] S1. Obtain a data set, which is composed of multiple groups of corresponding plant mesophyll conductance predictors at the leaf scale and plant mesophyll conductance at the leaf scale;

[0090] S2. Divide the data set into a training set and a test set;

[0091] S3. Use the training set data as input to train the random forest model, and adjust the model parameters according to the loss function during training to obtain a preliminary random forest model;

[0092] S4. Use the test set data as input to test the random forest model to obtain test results, calculate evaluation metrics based on the test results. If the evaluation metrics are worse than the set threshold, repeat S3 - S4 until the evaluation metrics are better than the set threshold to obtain a pre-trained random forest model.

[0093] Embodiment 2:

[0094] Based on Embodiment 1, this embodiment further provides a system for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence, including:

[0095] A memory for storing instructions.

[0096] A processor for executing the instructions, so that the system executes the method for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence as described in Embodiment 1.

[0097] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and deformations can be made, and these improvements and deformations should also be regarded as the protection scope of the present invention.

Claims

1. A method for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence, characterized in that, Including: Obtaining predictors of plant mesophyll conductance; Using the predictors of plant mesophyll conductance as inputs and outputting the plant mesophyll conductance based on a random forest model; Wherein, the predictors of plant mesophyll conductance include the open ratio of photosystem II reaction centers, the maximum photochemical quantum efficiency of photosystem II, atmospheric carbon dioxide concentration, leaf temperature, saturation vapor pressure deficit, and photosynthetically active radiation.

2. The method for estimating the mesophyll conductance of a plant using solar-induced chlorophyll fluorescence according to claim 1, wherein The maximum photochemical quantum efficiency of photosystem II is set as a constant; The atmospheric carbon dioxide concentration, leaf temperature, saturation vapor pressure deficit, and photosynthetically active radiation are obtained by measurement; The open ratio of photosystem II reaction centers is calculated through the relationship functions between solar-induced chlorophyll fluorescence and linear electron transfer rate and between linear electron transfer rate and the open ratio of photosystem II reaction centers, wherein the solar-induced chlorophyll fluorescence is obtained by measurement.

3. The method for estimating the mesophyll conductance of plants using solar-induced chlorophyll fluorescence according to claim 2, wherein The expression of the relationship function between the linear electron transfer rate and solar-induced chlorophyll fluorescence is: ； Among them, represents the linear electron transfer rate, represents the maximum photochemical quantum efficiency of photosystem II, , where represents the intrinsic thermal dissipation constant, represents the fluorescence rate constant, represents the open ratio of the reaction center of photosystem II, represents the sunlight-induced chlorophyll fluorescence.

4. The method for estimating the mesophyll conductance of plants using solar-induced chlorophyll fluorescence according to claim 2, characterized in that, The relationship function between the linear electron transfer rate and the open ratio of photosystem II reaction centers is obtained by fitting the open-close redox model of photosynthetic electron transfer using light response curves and CO2 response curves, and its expression is: ； ； ； ； Among them, represents the linear electron transfer rate, U represents the maximum oxidation potential of the cytochrome b6f complex for the plastoquinone and plastohydroquinone pools, represents the standardized temperature response function described by the redox reaction rate, represents the S-shaped function of the photosynthetically active radiation absorbed by the leaf, represents the redox balance function between photosystem II and cytochrome, represents the proportion of the reversible photosystem II reaction center, represents the open ratio of the photosystem II reaction center, represents the primary resistance of photosystem II electron transfer, represents the secondary resistance of photosystem II electron transfer, represents the standard reference temperature, represents the leaf temperature, represents the comprehensive temperature sensitivity parameter related to the Gibbs free activation energy, represents the stoichiometric parameter, 、 represent empirical parameters.

5. The method for estimating the mesophyll conductance of a plant using solar-induced chlorophyll fluorescence according to claim 4, wherein The unknown relationship parameters in the relationship function between the linear electron transfer rate and the open ratio of photosystem II reaction centers include the maximum oxidation potential of cytochrome b6f complex for plastoquinone and plastohydroquinone pools, the proportion of reversible photosystem II reaction centers, the primary resistance of photosystem II electron transfer, the secondary resistance of photosystem II electron transfer, the comprehensive temperature sensitivity parameter related to Gibbs free activation energy, stoichiometric parameters, and empirical parameters; The determination method of the unknown relationship parameters includes: Obtaining multiple sets of corresponding true values of linear electron transfer rate and true values of the open ratio of photosystem II reaction centers; Taking the multiple sets of true values of the open ratio of photosystem II reaction centers as inputs in sequence, and predicting the linear electron transfer rate through the relationship function between the linear electron transfer rate and the open ratio of photosystem II reaction centers combined with an optimization algorithm to obtain predicted values of the linear electron transfer rate; Calculating an objective function based on the predicted values of the linear electron transfer rate and the true values of the linear electron transfer rate, and the unknown relationship parameters corresponding to the minimum objective function value are the optimal solutions.

6. The method for estimating the mesophyll conductance of plants using solar-induced chlorophyll fluorescence according to claim 5, characterized in that, The optimization algorithm includes one or more of differential evolution optimization algorithm, basin hopping optimization algorithm, global optimization algorithm, dual annealing optimization algorithm, and DIRECT optimization algorithm.

7. The method for estimating the mesophyll conductance of a plant using solar-induced chlorophyll fluorescence according to claim 5, wherein The objective function is the sum of squared errors between the predicted values of the linear electron transfer rate and the true values of the linear electron transfer rate.

8. The method for estimating the mesophyll conductance of plants using solar-induced chlorophyll fluorescence according to claim 1, wherein The determination method of the predictors of plant mesophyll conductance includes: According to the variable J method model, the relationship factors of mesophyll conductance include net photosynthetic rate, linear electron transfer rate, leaf temperature, and intercellular carbon dioxide concentration; According to the photosynthesis physiological and biochemical model, the relationship factor of net photosynthetic rate is the maximum carboxylation efficiency, and the maximum carboxylation efficiency is estimated from the proportion of open reaction centers in photosystem II and solar-induced chlorophyll fluorescence; The relationship factors for the linear electron transfer rate obtained according to the MLR-SIF model are the open ratio of the photosystem II reaction center, the sunlight-induced chlorophyll fluorescence emitted by photosystem II, and the maximum photochemical quantum efficiency of photosystem II. Among them, the sunlight-induced chlorophyll fluorescence emitted by photosystem II is a function of the photosynthetically active radiation; The relationship factors for the intercellular carbon dioxide concentration obtained according to Fick's first law are the atmospheric carbon dioxide concentration, the net photosynthetic rate, and the stomatal conductance. Among them, the stomatal conductance is modeled based on the proportion of open reaction centers in photosystem II, the atmospheric carbon dioxide concentration, and the vapor pressure deficit; Combining all the relationship factors gives the plant mesophyll conductance predictor.

9. The method for estimating the mesophyll conductance of a plant using solar-induced chlorophyll fluorescence according to claim 1, wherein It also includes pre-training the random forest model, and the pre-training method includes: S1. Obtain a data set, which consists of multiple groups of corresponding plant mesophyll conductance predictors at the leaf scale and plant mesophyll conductance at the leaf scale; S2. Divide the data set into a training set and a test set; S3. Use the training set data as input to train the random forest model, and adjust the model parameters according to the loss function during training to obtain a preliminary random forest model; S4. Use the test set data as input to test the random forest model to obtain a test result, calculate the evaluation index according to the test result. If the evaluation index is worse than the set threshold, repeat S3 - S4 until the evaluation index is better than the set threshold to obtain a pre-trained random forest model.

10. A system for estimating plant mesophyll conductance using solar-induced chlorophyll fluorescence, characterized in that, It includes: A memory for storing instructions; A processor for executing the instructions, so that the system executes the method for estimating plant mesophyll conductance using sunlight-induced chlorophyll fluorescence as described in any one of claims 1 - 9.

Images