Physio-environment synergistically driven dynamic estimation method and system for isoprene emission

Abstract

The present application belongs to the technical field of biological volatile organic compound emission calculation, and discloses a physiological-environmental synergistically driven isoprene emission dynamic estimation method and system; the method comprises the following steps: S1, constructing a leaf scale emission model; S2, calculating substrate supply rate; S3, constructing a comprehensive temperature marginal response function and a carbon dioxide direct biochemical inhibition function to represent enzyme catalytic reaction efficiency; S4, obtaining multi-source research data sets of field observation, open-top box and sample plot control experiment, and adopting a two-stage hierarchical Bayesian method to estimate parameters of the comprehensive temperature marginal response function in step S3; S5, calculating leaf scale emission rate; S6, correcting the estimation in step S4. The present application realizes separation of the calculation framework of photosynthetic substrate production process and enzyme catalytic reaction process, overcomes inherent defects that the existing model mechanism path effect is easy to repeat and superimpose, and supports localization correction and uncertainty quantitative evaluation by using multi-source data.

Description

Technical Field

[0001] This invention relates to the field of biological volatile organic compound emission calculation technology, and in particular to a method and system for dynamic estimation of isoprene emissions driven by physiological and environmental factors. Background Technology

[0002] Isoprene is currently the most abundant biogenic volatile organic compound emitted in nature and is also an important precursor to the formation of tropospheric ozone and secondary organic aerosols. Accurate estimation of vegetation isoprene emission fluxes provides fundamental support for the development of atmospheric chemical transport models, Earth system models, and air quality early warning systems.

[0003] Isoprene is synthesized in chloroplasts via the methyl erythritol phosphate pathway. Its biosynthetic rate is jointly regulated by two types of biochemical mechanisms: its substrate supply rate is driven by the electron transport chain of photosynthesis and is affected by various meteorological environmental factors, including light; and the rate of enzyme catalytic reaction during its synthesis is affected by the content and activity of various enzymes, exhibiting typical enzyme kinetic characteristics.

[0004] Existing mainstream isoprene emission models are mainly divided into two categories, both of which have inherent defects:

[0005] The first category consists of empirical models, represented by the Model of Emissions of Gases and Aerosols from Nature (MEGAN). These models express emission rates as the product of a basic emission factor and multiple independent environmental correction coefficients. Their shortcomings include: a lack of a unified biochemical mechanism basis; temperature response parameters largely derived from controlled-environment experiments on seedlings, potentially leading to systematic biases compared to naturally grown adult plants; and ambiguous treatment of the effects of carbon dioxide, failing to distinguish between different pathways influencing substrate supply and enzyme activity.

[0006] The second category is semi-empirical photosynthetic coupling models, represented by the photosynthesis-based biogenic volatile organic compound (PS_BVOC) emission model. These models directly couple emission rates with photosynthetic rates. Their drawbacks include: in terms of temperature response, they do not reflect the inactivation characteristics of isoprene synthesis-related enzymes at high temperatures; high-temperature inhibition is achieved solely through changes in the carbon partition coefficient in substrate allocation; and in terms of carbon dioxide response, there may be redundant calculations of substrate supply pathway effects.

[0007] Furthermore, the parameters of the above schemes are usually based on limited laboratory data and lack systematic integration of global field observation data. The representativeness and uncertainty of the parameters have not been effectively quantified.

[0008] In summary, existing technologies fail to provide a solution that can simultaneously achieve the following objectives: based on plant physiological and biochemical research, separating the two types of mechanisms related to substrate supply and enzyme catalysis in a computational framework; independently modeling the differential effects of environmental factors on the two types of mechanisms; utilizing multi-source field observation and control experimental data to perform parameter estimation and uncertainty quantification using rigorous statistical methods; and supporting automatic localization correction of parameters directly using regional observation data in the model. Summary of the Invention

[0009] To address the shortcomings of existing technologies in simultaneously separating substrate supply and enzyme-catalyzed reactions within a computational framework, independently modeling the differential effects of meteorological environmental factors such as light, carbon dioxide, and temperature on these two mechanisms, and facilitating localized model calibration, this invention proposes a physiological-environment-driven dynamic estimation method and system for isoprene emissions. This method constructs emission models based on relevant plant physiological and biochemical mechanisms, employs hierarchical modeling based on multi-source field observations and control experiments, uses rigorous statistical methods for parameter estimation and uncertainty quantification, and supports direct localized parameter calibration using observation data from multiple regions.

[0010] This invention is achieved through the following technical solution, including the following steps:

[0011] S1. Based on the substrate supply rate and the enzyme catalytic reaction rate during isoprene synthesis, construct the leaf-scale instantaneous isoprene emission rate;

[0012] S2. Construct the substrate supply rate based on the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient;

[0013] S3. Based on the comprehensive temperature marginal response function and the direct biochemical inhibition function of carbon dioxide, construct the enzyme catalytic reaction rate for isoprene synthesis;

[0014] S4. Obtain field observation, open-top box or plot control experiment datasets from multiple research sources, and use a two-stage hierarchical Bayesian method to estimate the parameters in the comprehensive temperature marginal response function in step S3, and obtain the global mean and posterior probability distribution of the parameters.

[0015] S5. Based on the instantaneous isoprene emission rate at the leaf scale obtained in step S1, the light-limited photosynthetic rate and photosynthetic carbon allocation coefficient obtained in step S2, the comprehensive temperature marginal response function and carbon dioxide direct inhibition function obtained in step S3, and the global mean and posterior probability distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4, the instantaneous isoprene emission rate at the leaf scale is obtained.

[0016] As a further preferred embodiment, the formula for the substrate supply rate in step S2 is as follows:

[0017] S=A iso ×Y α ;

[0018] In the formula, S represents the substrate supply rate; A iso Indicates the light-limited photosynthetic rate; Y α This represents the photosynthetic carbon partition coefficient.

[0019] As a further preferred option, the specific steps of step S3 are as follows:

[0020] S3.1. The comprehensive temperature marginal response function is constructed using a logarithmic space expansion based on the thermodynamic equilibrium of enzyme protein folding-unfolding, as shown in the following formula:

[0021] ;

[0022] In the formula, This represents the combined temperature marginal response function; Represents a parameter vector. This is obtained by a two-stage hierarchical Bayesian method; Indicates the blade temperature; This indicates the temperature under standard conditions. Indicates the activation enthalpy on the low-temperature side; This indicates the enthalpy change of enzyme thermal denaturation and unfolding. R represents the entropy change during enzyme thermal denaturation and unfolding; R represents the universal gas constant. This represents the log-scale magnitude parameter, obtained by hierarchical Bayes estimation;

[0023] S3.2. Based on the normalized function of the total effect of carbon dioxide and the normalized ratio of the carbon dioxide effect, the direct biochemical inhibition function of carbon dioxide is obtained.

[0024] As a further preferred option, the specific steps of step S4 are as follows:

[0025] S4.1, Stage 1: Parameter estimation in the comprehensive temperature marginal response function:

[0026] Using the full dataset, a two-stage, three-layer hierarchical Bayesian model is constructed. Markov chain Monte Carlo sampling is performed using a decentralized parameterization scheme. After convergence, the global mean and posterior distribution of the parameters of the comprehensive temperature marginal response function are obtained.

[0027] S4.2, Stage Two: Using a subset of the dataset containing measured illumination data, the posterior distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4.1 is used as the prior information to estimate the illumination saturation response parameters. An observation layer is constructed using the illumination saturation response parameters for posterior sampling. After convergence, the final global mean and posterior probability distribution of the parameters of the comprehensive temperature marginal response function are output.

[0028] As a further preferred embodiment, the calculation formula for step S5 is as follows:

[0029] ;

[0030] In the formula, This represents the instantaneous emission rate of isoprene at the blade scale; These represent the emission factors for each vegetation functional type or tree species; This represents the normalized value of the substrate supply term under reference conditions.

[0031] This invention also provides a physiological-environment synergistically driven dynamic estimation system for isoprene emissions. This system is used to implement the physiological-environment synergistically driven dynamic estimation method for isoprene emissions described in this invention. The system includes:

[0032] The data input module is used to receive meteorological driving data, vegetation parameters, and control information;

[0033] The substrate supply calculation module is used to calculate the light-limited photosynthetic rate and photosynthetic carbon allocation coefficient;

[0034] The enzyme catalytic efficiency calculation module is used to perform calculations of the combined temperature marginal response function and the direct biochemical inhibition function of carbon dioxide;

[0035] The parameter estimation module is used to estimate the parameters in the comprehensive temperature marginal response function using a two-stage hierarchical Bayesian method.

[0036] The emission flux calculation module is used to calculate the instantaneous emission rate of isoprene at the leaf scale, and to perform canopy-scale integration by calling the canopy module of external dynamic vegetation model or atmospheric chemistry model, supporting atmospheric chemistry impact analysis.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] 1. This invention decomposes the leaf-scale isoprene emission rate into a substrate supply module (driven by a photosynthetic biochemical model) and an enzyme catalytic reaction efficiency module (composed of a combined temperature marginal response function and a direct carbon dioxide biochemical inhibition function). The two modules have independent biochemical mechanism foundations and replaceable parameterization schemes, realizing the separation of the computational framework of the photosynthetic substrate production process and the enzyme catalytic conversion process, and overcoming to some extent the inherent defects of existing model mechanism confusion.

[0039] 2. This invention achieves quantitative separation of two effect pathways of carbon dioxide on isoprene emissions (substrate supply indirect pathway and enzyme activity biochemical inhibition pathway), avoiding the double counting of carbon dioxide effects in the substrate supply pathway; at the same time, it adopts a mechanism-heuristic temperature response function based on the thermodynamic equilibrium of enzyme protein folding-unfolding, which has higher parameter freedom and thermodynamic interpretability among different vegetation functional species, and facilitates greater improvement of unreasonable predictions under extreme temperature conditions.

[0040] 3. This invention employs a two-stage hierarchical Bayesian model, which supports the simultaneous fusion of field observation data from multiple research sources as well as traditional control experimental data from open-top boxes, plots, or laboratory environments. This achieves joint constraints on multi-source heterogeneous data within a unified statistical inference framework. The resulting posterior distribution of parameters simultaneously quantifies parameter uncertainty, overcoming the inherent defects of representativeness bias and unassessable uncertainty in traditional fixed parameterization schemes, and providing stronger ecological representativeness for naturally grown mature plants.

[0041] 4. This invention provides a scheme for localized parameter correction using field observation data, which has good regional adaptability and scale expansion capability. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the overall calculation method and system of the present invention.

[0043] Figure 2 The carbon dioxide response fitting curve is obtained using the default global carbon dioxide control experimental database in this invention.

[0044] Figure 3 The fitting results of the two-stage hierarchical Bayesian model in the embodiments of the present invention are obtained using the default global isoprene observation database.

[0045] Figure 4 This is a comparison chart of model calculation and typical site observation in an embodiment of the present invention. Detailed Implementation

[0046] The advantages and features of the present invention will be illustrated and explained by the following non-limiting description of preferred embodiments, which are given by way of example only with reference to the accompanying drawings.

[0047] like Figure 1 As shown, this invention provides a method for dynamically estimating isoprene emissions driven by a synergistic physiological and environmental factor, comprising the following steps:

[0048] Step S1: Based on the substrate supply rate and the catalytic efficiency of isoprene synthase, construct the leaf-scale instantaneous isoprene emission rate, which serves as the calculation framework for isoprene emissions. The formula is as follows:

[0049] ;

[0050] In the formula, This represents the instantaneous emission rate of isoprene at the blade scale; Indicates the substrate supply rate; Indicates the catalytic efficiency of isoprene synthase; and The photosynthetic-driven flux of dimethylallyl pyrophosphate (DMAPP) and the rate of enzyme-catalyzed conversion of precursors to isoprene in the 2-C-methyl-D-erythritol 4-phosphate (MEP) pathway, respectively, exhibit independent and clear biochemical correspondences. In this invention, the substrate supply rate... The catalytic efficiency of isoprene synthase was calculated by the substrate supply calculation module. The efficiency is calculated by the enzyme catalysis efficiency calculation module. Both the substrate supply calculation module and the enzyme catalysis efficiency calculation module can be independently parameterized and replaced, and have modularity and replaceability.

[0051] Step S2: Construct the substrate supply rate based on the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient, as shown in the following formula:

[0052] ;

[0053] In the formula, Indicates the substrate supply rate; The light-limited photosynthetic rate is obtained by solving the Farquhar-von Caemmerer-Berry biochemical model of photosynthesis (FvCB model) and the Ball–Berry model of stomatal conductance (Ball–Berry model). It is the sum of the net light-limited assimilation rate and the dark respiration rate. The photosynthetic carbon allocation coefficient is determined by the algebraic relationship between the carbon dioxide concentration between leaf cells and the carbon dioxide compensation point, and characterizes the proportion of DMAPP synthesis allocated to a unit of photosynthetic electron transport flux.

[0054] In this invention, the light-limited photosynthetic rate is obtained by simultaneously solving the FvCB photosynthetic biochemical model (von Caemmerer S et al., “A perspective: A biochemical model of photosynthetic CO2 assimilation in leaves of C3species.” Planta, 2025, 262: 115; DOI: 10.1007 / s00425-025-04834-7) and the Ball-Berry stomatal conductance model (Ball JT et al., “A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions.” In: Biggins J. (eds), Progress in Photosynthesis Research, 1987:221-224; DOI: 10.1007 / 978-94-017-0519-6_48). The photosynthetic carbon allocation coefficient is derived from this algebraic expression based on the stoichiometry of carbon metabolism. Both are existing technologies, and will not be described further in this invention.

[0055] Complete calculations rely on meteorological variables such as leaf temperature, photosynthetically active radiation, atmospheric carbon dioxide concentration, atmospheric pressure, and relative humidity, as well as vegetation parameters such as maximum carboxylation rate and quantum efficiency. The impact of water stress on photosynthetic substrates is reflected in intercellular carbon dioxide concentration and... middle. The carbon dioxide photosynthetic compensation point, which is dependent on both intercellular carbon dioxide concentration and temperature-regulated, is a key intermediate in separating the carbon dioxide effect pathway in step S3, characterizing the indirect effect of carbon dioxide in the substrate supply pathway.

[0056] Step S3: Construct the catalytic efficiency of isoprene synthase based on the comprehensive temperature marginal response function and the direct biochemical inhibition function of carbon dioxide;

[0057] Step S3.1: Construct the comprehensive temperature marginal response function using the logarithmic space expansion form based on the thermodynamic equilibrium of enzyme protein folding-unfolding, as shown in the following formula:

[0058] ;

[0059] In the formula, This represents the combined temperature marginal response function; Represents a parameter vector. This is obtained by a two-stage hierarchical Bayesian method; This indicates the blade temperature, expressed in Kelvin (K). This represents the temperature under standard conditions, in Kelvin (K). The value is 303.15K, which is consistent with the standard emission factor. The measurement conditions were consistent; This represents the activation enthalpy on the low-temperature side, in units of... This characterizes the temperature sensitivity in the peak Arrhenius temperature response function of enzyme-catalyzed reactions; This indicates the enthalpy change of enzyme during thermal denaturation and unfolding, measured in units of... Characterizes the enthalpy driving force of enzyme protein unfolding reaction; This represents the entropy change during enzyme thermal denaturation and unfolding, measured in units of... , representing the entropy driving force of the unfolding reaction of enzyme proteins; R represents the universal gas constant, which is 8.314. ; This represents the logarithmic scaling magnitude parameter, obtained by hierarchical Bayesian estimation.

[0060] The parameter vector in this step A two-stage hierarchical Bayesian method is used to statistically estimate the global mean and posterior probability distribution of parameters from observational data from multiple studies. When data is sufficient, this method allows independent changes in the enthalpy and entropy of enzyme thermostability for different vegetation functional types, reflecting the differences in the thermodynamic mechanisms of enzyme protein unfolding.

[0061] The comprehensive temperature marginal response function adopts a logarithmic space expansion form based on the thermodynamic equilibrium of enzyme protein folding-unfolding, and includes three independent parameters: activation enthalpy on the low-temperature side, unfolding enthalpy change on the high-temperature side, and unfolding entropy change. The function monotonically increases with increasing temperature when the leaf temperature is below the peak temperature, and then rapidly decreases due to the thermal denaturation and inactivation of enzyme protein. The parameters are statistically estimated from multiple study observation data using a two-stage hierarchical Bayesian method, reflecting the comprehensive observational marginal effect of temperature dependence of enzyme catalytic activity after removing the influence of photosynthetic substrate supply.

[0062] Since the substrate supply module already incorporates the photosynthetic system's response to temperature within the FvCB model, the combined temperature marginal response function... It primarily captures the residual temperature response outside the substrate supply module, which mainly corresponds to the change in enzyme catalytic activity with temperature. It can also absorb the residual temperature error of the substrate supply model, giving the model a certain degree of adaptive correction capability.

[0063] Step S3.2: Based on the normalized function of the total effect of carbon dioxide and the normalized ratio of the carbon dioxide effect, obtain the direct biochemical inhibition function of carbon dioxide.

[0064] Step S3.2.1: Based on multiple carbon dioxide concentration gradient control data, the normalized function of the total carbon dioxide effect is obtained by grouping and fitting, as shown in the following formula:

[0065] ;

[0066] In the formula, This represents the normalized function for the total effect of carbon dioxide; This indicates the concentration of carbon dioxide in the atmosphere; Indicates the carbon dioxide concentration under standard conditions (in this invention) ); and All of these are fitting parameters obtained from carbon dioxide control experimental data, with default values ​​of 0.3711, 0.002792 and 0.6289, respectively.

[0067] Step S3.2.2: Based on the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient, construct the normalized ratio of the carbon dioxide effect, as shown in the following formula:

[0068] ;

[0069] In the formula, This represents the normalized ratio of the carbon dioxide effect, which is the normalized ratio of the carbon dioxide effect in the substrate supply pathway. It is used to quantitatively characterize the indirect carbon dioxide effect already included in the substrate supply calculation module. This refers to the carbon dioxide concentration in leaf cells. This represents the concentration of carbon dioxide in leaf cells under standard conditions, and , This refers to the carbon dioxide translocation ratio of the leaves, with a typical value of approximately 0.7. This is the carbon dioxide compensation point; This indicates that the photosynthetic rate is light-limited. Indicates the photosynthetic carbon partition coefficient; and These represent the standard state ( ,in The light-limited photosynthetic rate and the isoprene synthesis partition coefficient for photosynthetically produced carbon under standard radiation intensity are both obtained by the method in step S2.

[0070] The normalized ratio of the carbon dioxide effect in this step , which can be determined by The algebraic relationship can be calculated directly, and the light-related quantities will be eliminated by simplification in the ratio, so actual light data is not required when fitting.

[0071] Step S3.2.3: Multiply the normalized function of the total carbon dioxide effect by the normalized ratio of the carbon dioxide effect to obtain the direct biochemical inhibition function of carbon dioxide, as shown in the following formula:

[0072] ;

[0073] In the formula, This represents the direct biochemical inhibition function of carbon dioxide.

[0074] The overall approach to constructing the direct biochemical inhibition function of carbon dioxide in this invention is as follows: The comprehensive effect of carbon dioxide on isoprene can be decomposed into an indirect path that regulates substrate supply by changing intercellular carbon dioxide concentration and a direct path that directly inhibits key enzymatic reactions in the MEP pathway. In this method, the indirect path is provided by the substrate supply module, and the effects of the direct path are separated as follows: First, based on multiple carbon dioxide concentration gradient control experimental data, a normalized function of the total carbon dioxide effect is obtained by group fitting. This function simultaneously includes both the indirect carbon dioxide effect of the substrate supply pathway and the direct biochemical inhibitory effect of related enzyme reactions, which need to be further separated. Then, based on the mathematical model of the substrate supply term, the normalized ratio of the carbon dioxide effect of the substrate supply pathway is calculated. This ratio is used to quantitatively characterize the indirect carbon dioxide effect already included in the substrate supply module; it can be obtained from... The algebraic relationship is directly calculated, where light-related quantities are eliminated by simplification in the ratio, eliminating the need for measured light data. Finally, the influence of the substrate supply pathway is removed, yielding a function of carbon dioxide concentration directly inhibiting enzyme-related reactions. Through this pathway separation process, By retaining only the direct biochemical inhibitory effect of carbon dioxide on the key enzymatic steps of the MEP pathway and eliminating the redundant inclusion of the carbon dioxide response in the indirect pathway of substrate supply, the two types of effects are quantitatively separated, giving the emission response under the scenario of increased carbon dioxide a clear mechanistic explanation.

[0075] Step S4: Obtain field observation, open-top box, or sample plot control experiment datasets from multiple research sources. Use a two-stage hierarchical Bayesian method to estimate the parameters in the comprehensive temperature marginal response function from step S3, and obtain the global mean and posterior probability distribution of the parameters.

[0076] By default, the dataset used in this step is based on publicly available global observation data at a global scale. When applied to a specific region, the parameters can be estimated again using local continuous observation data from that region.

[0077] Step S4.1, Stage 1: Parameter estimation in the comprehensive temperature marginal response function:

[0078] Using the full dataset, a three-layer hierarchical Bayesian model is constructed. Markov Chain Monte Carlo (MCMC) sampling is performed using a decentralized parameterization scheme. After convergence, the posterior distribution of the parameters of the comprehensive temperature marginal response function is obtained. The three-layer hierarchical Bayesian model is as follows:

[0079] Hyperparameter layer: ;

[0080] Research level: ;

[0081] Grouping layer: ;

[0082] Observation layer: ;

[0083] In the formula, is the mean vector of global hyperparameters, representing the overall mean of the parameters of the comprehensive temperature marginal response function over all studies and groups; To provide reasonable constraints on the prior distribution information of hyperparameters when strong prior information is lacking, and to avoid excessive divergence in estimation; The global hyperparameter covariance matrix describes the prior uncertainty and related structure among the components of the parameter vector. This represents the random effect of the s-th study, i.e., the offset of the study stratum, which indicates the systematic offset caused by differences in experimental conditions, measurement methods, etc. express It follows a mean of 0 and a variance of . The normal distribution; The variance between studies reflects the degree of dispersion of random effects among different studies; Number the research index; This represents the total number of studies included in the model. This represents the parameter vector in the comprehensive temperature marginal response function of group g; This indicates the random effects in the study stratum of group g in study s; The covariance matrix at the g level represents the uncertainty of parameters within that group; G represents the total number of groups. This represents the emission rate of the kth observation; This represents the log-scale amplitude parameters of the grouped layers, including variations in photosynthesis and other amplitude differences. This represents the parameter vector in the comprehensive temperature marginal response function of the group g to which the kth observation belongs; This represents the temperature of the k-th observation; Indicates the systematic bias correction coefficient for the experiment type; Indicator variables representing the type of experiment (characterizing ordinary observation or control experiments); Variance represents the variance of the observation layer, reflecting the random fluctuations between the model's predicted values ​​and the actual observed values ​​caused by factors such as measurement errors.

[0084] The impact of changes in illumination on emission amplitude is determined by stratified amplitude parameters. The interlayer variation is naturally absorbed, and the estimated combined temperature marginal response parameter reflects the marginal effect of temperature observation after light is marginalized. The parameter reflects the observed marginal temperature response under the combined effect of the above mechanisms, and is not directly equivalent to the parameters determined by in vitro enzyme kinetic experiments of a single synthase, but is compatible with the reported thermostability parameters of plant isoprene synthase in terms of magnitude and trend.

[0085] Step S4.2, Stage Two: Using a subset of the dataset containing measured illumination data, the posterior distribution of the parameters of the comprehensive temperature marginal response function obtained in Step S4.1 is used as the prior information to estimate the illumination saturation response parameters. An observation layer is constructed using the illumination saturation response parameters for posterior sampling. After convergence, the final posterior probability distribution of the parameters of the comprehensive temperature marginal response function is output.

[0086] Step S4.2.1: Construct the light response function, as shown in the following formula:

[0087] ;

[0088] In the formula, Represents the light response function; This represents the light intensity (effective photosynthetic radiation) of the kth observation. This represents the light intensity under standard conditions (in this study) ), These are the fitting parameters obtained using the observed data.

[0089] Step S4.2.2: Based on the illumination response function, a subset of the dataset containing measured illumination data is used, and the posterior distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4.1 is used as the prior information to estimate the illumination saturation response parameters.

[0090] Step S4.2.3: Construct an observation layer using parameters that include the saturated illumination response for posterior sampling. After convergence, output the final posterior probability distribution of the parameters of the comprehensive temperature marginal response function. The formula for constructing the observation layer using illumination saturation response parameters is as follows:

[0091] ;

[0092] Preferably, in this invention, both stages of the two-stage hierarchical Bayesian parameter estimation employ a No-U-Turn Sampler (NUTS) for posterior sampling. A preferred configuration is as follows: four parallel chains, a 3000-step warm-up, and 3000 sampling steps per chain, using the Gelman-Rubin Convergence Diagnostic Statistic (Gelman-Rubin). This serves as a convergence criterion. The final output includes the global mean and posterior probability distribution of each parameter, which are used for emission rate calculation and uncertainty analysis.

[0093] The purpose of this step is to remove the influence of changes in photosynthetic products on the temperature function.

[0094] like Figure 2 The figure shows the standardized function curve of the total carbon dioxide effect obtained by fitting data on the impact of globally published carbon dioxide control experiments on the isoprene emission rate of vegetation using the estimation method of this invention. Different scatter points represent data from independent studies, with emissions at a carbon dioxide concentration of 400 ppm as the baseline, and each study was independently standardized. Bins are divided at 25 ppm carbon dioxide concentration intervals; the bin length shows the interquartile range of the data within that range, the whisker line shows the range of 1.5 times the interquartile range, and the black line within the bin shows the median of the data within that range. The black fitted curve represents the currently used default fitting relationship, which is derived from the medians of each bin, but excludes two abnormally high-temperature-promoted data points that are inconsistent with current mechanistic studies. The fitted values ​​of each parameter in the fitted curve are indicated in the upper right corner.

[0095] like Figure 3As shown, a two-stage hierarchical Bayesian model was driven by publicly available global vegetation isoprene emission observation data to obtain the joint posterior distribution of parameters for the comprehensive temperature marginal response function. The model has converged stably and serves as the default parameter source in the current method. From left to right, these represent activation energy, inactivation enthalpy, and high-temperature suppression entropy change, respectively. In each figure, the thick solid line inside the box represents the median, the white rhombus represents the mean, the solid whisker line extends to the farthest actual data point within 1.5 times the interquartile range, and the dashed whisker line shows the 89% highest density interval (model confidence interval). The light gray scatter dots behind the box represent sampled values, and the hollow circles represent outlier posterior samples falling outside the whisker line range. The mean, median, and standard deviation of the parameter are labeled in the upper right corner of each subfigure.

[0096] Step S5: Based on the leaf-scale instantaneous isoprene emission rate from Step S1, the photosynthetic rate under light limitation from Step S2, the partition coefficient of photosynthetically generated carbon in isoprene synthesis, the comprehensive temperature marginal response function from Step S3 (with parameters provided by the global mean from Step S4), and the direct biochemical inhibition function of carbon dioxide, the leaf-scale instantaneous isoprene emission rate is obtained, as shown in the following formula:

[0097] ;

[0098] In the formula, This represents the instantaneous emission rate of isoprene at the blade scale; These represent the emission factors for each vegetation functional type or tree species; This represents the normalized value of the substrate supply term under reference conditions, making the model's output equal to the reference value under standard conditions. .

[0099] After obtaining the instantaneous isoprene emission rate at the leaf scale, this invention can calculate the emission flux at the vegetation scale through canopy integration. The canopy integration related calculations can be achieved by calling the canopy module of an external dynamic vegetation model or atmospheric chemistry model, and are not limited to a specific software form.

[0100] Preferably, in some embodiments, step S5 is followed by step S6, which is as follows: using continuous field observation data of the target area, and with the global-scale posterior distribution obtained in step S4 as the prior, the comprehensive temperature marginal response parameters are locally re-estimated using a Bayesian method, and the measured emission factors can be used for local species or vegetation communities. This enables in-situ parameter correction, reflecting the actual temperature adaptability of vegetation in the target area.

[0101] Field calibration focuses on the comprehensive temperature marginal response parameter for the following reasons: Temperature is the largest controlling factor for isoprene emission rates under natural conditions during periods of light; the comprehensive temperature marginal response parameter can also be robustly constrained by continuous flux observation data in the field; and data from mature plants in the field can compensate for the systematic bias introduced by traditional open-top chamber or growth chamber experiments, which mostly only measure seedlings, resulting in parameters that better reflect the actual temperature sensitivity of mature plants under natural growth conditions. Currently, the direct biochemical inhibition parameter of carbon dioxide is derived from publicly available global carbon dioxide concentration gradient control experimental data. The magnitude of carbon dioxide changes in current regional continuous field observations is usually insufficient to effectively constrain this parameter, and local re-estimation is generally unnecessary unless there is a specific requirement.

[0102] like Figure 4 The chart shows a comparison between publicly available site observation data and simulations performed using the current calculation method with default parameters. Solid circles (solid lines) represent observed site values, while hollow triangles (dashed lines) represent simulated values. The x-axis displays the observed plant species at each site, and the subplots display the coefficient of determination (COP) for that site. The results of the calculation method and the root mean square error (RMSE) show that the method can effectively simulate the isoprene emissions of different vegetation species in different regions.

[0103] In some embodiments, the present invention also provides a system for realizing the physiological-environment synergistically driven dynamic estimation of isoprene emissions as described in the present invention, comprising:

[0104] The data input module is used to receive meteorological driving data, vegetation parameters, and control information;

[0105] The substrate supply calculation module is used to calculate the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient;

[0106] The enzyme catalytic efficiency calculation module is used to perform calculations of the combined temperature marginal response function and the direct biochemical inhibition function of carbon dioxide;

[0107] The parameter estimation module is used to estimate the parameters in the comprehensive temperature marginal response function using a two-stage hierarchical Bayesian method.

[0108] The emission flux calculation module is used to calculate the instantaneous emission rate of isoprene at the leaf scale, and can achieve canopy-scale integration by calling the canopy module of external dynamic vegetation model or atmospheric chemistry model, supporting atmospheric chemistry impact analysis.

[0109] In addition to the above embodiments, the present invention may have other implementation methods. All technical solutions formed by equivalent substitution or equivalent transformation fall within the protection scope claimed by the present invention.

Claims

1. A method for dynamically estimating isoprene emissions driven by a physiological-environmental synergy, characterized in that: Includes the following steps: S1. Based on the substrate supply rate and the catalytic efficiency of isoprene synthase, the instantaneous isoprene emission rate at the leaf scale was constructed. S2. Construct the substrate supply rate based on the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient; S3. Construct the catalytic efficiency of isoprene synthase based on the comprehensive temperature marginal response function and the direct biochemical inhibition function of carbon dioxide. S4. Obtain field observation, open-top box or plot control experiment datasets from multiple research sources, and use a two-stage hierarchical Bayesian method to estimate the parameters in the comprehensive temperature marginal response function in step S3, and obtain the global mean and posterior probability distribution of the parameters. S5. Based on the instantaneous isoprene emission rate at the leaf scale obtained in step S1, the light-limited photosynthetic rate and photosynthetic carbon allocation coefficient obtained in step S2, the comprehensive temperature marginal response function and carbon dioxide direct inhibition function obtained in step S3, and the global mean and posterior probability distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4, the instantaneous isoprene emission rate at the leaf scale is obtained.

2. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 1, characterized in that: The formula for the substrate supply rate in step S2 is as follows: S=A iso ×Y α ; In the formula, S represents the substrate supply rate; A iso Indicates the light-limited photosynthetic rate; Y α This represents the photosynthetic carbon partition coefficient.

3. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 2, characterized in that: The specific steps of step S3 are as follows: S3.

1. The comprehensive temperature marginal response function is constructed using a logarithmic space expansion based on the thermodynamic equilibrium of enzyme protein folding-unfolding, as shown in the following formula: ； In the formula, This represents the combined temperature marginal response function; Represents a parameter vector. This is obtained by a two-stage hierarchical Bayesian method; Indicates the blade temperature; This indicates the temperature under standard conditions. Indicates the activation enthalpy on the low-temperature side; This indicates the enthalpy change of enzyme thermal denaturation and unfolding. R represents the entropy change during enzyme thermal denaturation and unfolding; R represents the universal gas constant. This represents the log-scale magnitude parameter, obtained by hierarchical Bayes estimation; S3.

2. Based on the normalized function of the total effect of carbon dioxide and the normalized ratio of the carbon dioxide effect, the direct biochemical inhibition function of carbon dioxide is obtained.

4. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 3, characterized in that: The specific steps of step S3.2 are as follows: S3.2.1 Based on multiple carbon dioxide concentration gradient control data, the normalized function of the total carbon dioxide effect is obtained by grouping and fitting, as shown in the following formula: ； In the formula, This represents the normalized function for the total effect of carbon dioxide; This indicates the concentration of carbon dioxide in the atmosphere; This indicates the carbon dioxide concentration under standard conditions. and All parameters are fitted parameters obtained from carbon dioxide control experimental data; S3.2.2 Based on the light-limited photosynthetic rate and the photosynthetic carbon allocation coefficient, construct the normalized ratio of the carbon dioxide effect, as shown in the following formula: ； In the formula, This represents the normalized ratio of the carbon dioxide effect; This refers to the carbon dioxide concentration in leaf cells. This represents the concentration of carbon dioxide in leaf cells under standard conditions, and , The carbon dioxide transfer ratio of the leaves; This is the carbon dioxide compensation point; Indicates the photosynthetic rate limited by light; Indicates the photosynthetic carbon partition coefficient; and represents the light-limited photosynthetic rate and the isoprene synthesis partition coefficient for photosynthetic carbon production under standard conditions, respectively. , , and All are obtained from step S2; S3.2.3 Multiply the normalized function of the total carbon dioxide effect by the normalized ratio of the carbon dioxide effect to obtain the direct biochemical inhibition function of carbon dioxide, as shown in the following formula: ； In the formula, This represents the direct biochemical inhibition function of carbon dioxide.

5. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 4, characterized in that: The specific steps of step S4 are as follows: S4.1, Stage 1: Parameter estimation in the comprehensive temperature marginal response function: Using the full dataset, a three-layer hierarchical Bayesian model is constructed. Markov chain Monte Carlo sampling is performed using a decentralized parameterization scheme. After convergence, the posterior distribution of the parameters of the comprehensive temperature marginal response function is obtained. The three-layer hierarchical Bayesian model is as follows: Hyperparameter layer: ; Research level: ; Grouping layer: ; Observation layer: ; In the formula, Represents the mean vector of global hyperparameters; Represents the prior distribution information of hyperparameters; Represents the global hyperparameter covariance matrix; This represents the random effect of the s-th study; express It follows a mean of 0 and a variance of . The normal distribution; Indicates the variance between studies; Indicates the research index number; This represents the total number of studies included in the model; This represents the parameter vector in the comprehensive temperature marginal response function of group g; This indicates the random effects in the study stratum of group g in study s; Let G represent the covariance matrix at the g-level of grouping; G represents the total number of groups. This represents the emission rate of the kth observation; This represents the logarithmic scaling magnitude parameter of the grouping layer; This represents the parameter vector in the comprehensive temperature marginal response function of the group g to which the kth observation belongs; This represents the temperature of the k-th observation; Indicates the systematic bias correction coefficient for the experiment type; Indicator variable for experiment type; Indicates the variance of the observation layer; S4.

2. Using a subset of the dataset containing measured illumination data, the posterior distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4.1 is used as the prior information to estimate the illumination saturation response parameters. An observation layer is constructed using the illumination saturation response parameters for posterior sampling. After convergence, the final posterior probability distribution of the parameters of the comprehensive temperature marginal response function is output.

6. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 5, characterized in that: The specific steps of step S4.2 are as follows: S4.2.1 Construct the light response function, the formula is as follows: ； In the formula, Represents the light response function; This represents the effective photosynthetic radiation observed in the k-th observation; This represents the light intensity under standard conditions; This represents the fitting parameters based on the observed data; S4.2.

2. Based on the illumination response function, a subset of the dataset containing measured illumination data is used, and the posterior distribution of the parameters of the comprehensive temperature marginal response function obtained in step S4.1 is used as the prior information to estimate the corresponding parameters of the illumination response function. S4.2.

3. Construct the observation layer using parameters that include the saturated illumination response for posterior sampling. After convergence, output the final posterior probability distribution of the parameters of the comprehensive temperature marginal response function. The formula for constructing the observation layer using illumination saturation response parameters is as follows: 。 7. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 6, characterized in that: The calculation formula for step S5 is as follows: ； In the formula, This represents the instantaneous emission rate of isoprene at the blade scale; These represent the emission factors for each vegetation functional type or tree species; This represents the normalized value of the substrate supply term under reference conditions.

8. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to any one of claims 1 to 7, characterized in that: Step S5 may be followed by step S6: using continuous field observation data of the target area, with the global-scale posterior distribution obtained in step S4 as the prior, the parameters of the integrated temperature marginal response function are locally re-estimated using the Bayesian method, and the measured emission factors can be used for local species or vegetation communities to achieve field correction of the parameters.

9. The method for dynamically estimating isoprene emissions driven by physiological-environmental synergy according to claim 6, characterized in that: In step S4, both stages of the two-stage hierarchical Bayes parameter estimation use a non-rotating sampler for posterior sampling.

10. A physiological-environment synergistically driven dynamic estimation system for isoprene emissions, characterized in that: This system is used to implement the physiological-environment synergistically driven dynamic estimation method for isoprene emissions as described in any one of claims 1 to 9, and the system comprises: The data input module is used to receive meteorological driving data, vegetation parameters, and control information; The substrate supply calculation module is used to calculate the light-limited photosynthetic rate and photosynthetic carbon allocation coefficient; The enzyme catalytic efficiency calculation module is used to perform calculations of the combined temperature marginal response function and the direct biochemical inhibition function of carbon dioxide; The parameter estimation module is used to estimate the parameters in the comprehensive temperature marginal response function using a two-stage hierarchical Bayesian method. The emission flux calculation module is used to calculate the instantaneous emission rate of isoprene at the leaf scale, and to perform canopy-scale integration by calling the canopy module of external dynamic vegetation model or atmospheric chemistry model, supporting atmospheric chemistry impact analysis.

Images