A method for estimating the productivity of global land and marine ecosystems based on SIF and light response

Abstract

The present invention provides a method for estimating the productivity of global land and sea ecosystems based on SIF and light response, including: Step 1, fusing satellite remote sensing image data and meteorological data and performing consistency verification; Step 2, at each pixel point, substituting the input remote sensing data into the mechanical light response (MLR) theoretical formula to estimate the gross primary productivity (GPP), calculating the carbon dioxide partial pressure in the chloroplast stroma, the carbon dioxide compensation point of the chloroplast, the open ratio of the reaction center of the photosynthetic system, the energy ratio of plant heat dissipation and plant-emitted fluorescence, and the escape rate; Step 3, obtaining the distributions of C3 plants and C4 plants using vegetation type distribution data; calculating the gross primary productivity of C3 plants and C4 plants respectively. This solution simplifies the required quantity of vegetation parameters during the inversion process, greatly reduces the difficulty of parameter acquisition while ensuring the inversion accuracy, and at the same time significantly reduces the result uncertainty caused by parameters.

Description

Technical Field

[0001] The present invention belongs to the technical field of production force quantification assessment, and particularly relates to a method for estimating the productivity of global land and sea ecosystems based on SIF and light response. Background Art

[0002] Chlorophyll fluorescence has been proven in recent years to be applicable to estimating gross primary productivity. Based on this discovery, chlorophyll fluorescence inversion technology has begun to be widely used in the fields of agriculture, forestry, and environmental monitoring, especially in the assessment of vegetation growth status. This technology constructs a method for estimating the gross primary productivity of vegetation that is common to various ecosystems on land and sea globally based on the recently proposed mechanism of light response theory and combined with remote sensing data.

[0003] Gross Primary Productivity (GPP), as an important metric for the photosynthesis rate of vegetation, its real-time monitoring and accurate estimation are of great significance for clarifying the impact of climate change and human activities on ecosystems. The model method is the main tool for calculating GPP in large regions and globally. However, due to differences in the structures, parameters, and input data of various models, the total amount and spatio-temporal distribution of global GPP calculated by different models vary greatly. This has caused great uncertainty in large-scale GPP estimation and further analysis of ecosystems.

[0004] Sun-induced Chl Fluorescence (SIF) is a special fluorescence emitted by excited chlorophyll molecules. As an energy expenditure, it shares the energy provided by photosynthetically active radiation with photochemical reactions and thermal dissipation. More importantly, it is tightly coupled with photosynthesis in basic biochemical and biophysical processes, which indicates that SIF can be used as a way to accurately estimate gross primary productivity. In addition, the physiological basis of the SIF-GPP coupling ensures that SIF can track changes in photosynthetic rate under different environmental conditions. Therefore, the possibility of simply and accurately calculating GPP using chlorophyll fluorescence has currently received increasing attention.

[0005] Studies based on ground, airborne, and satellite data have shown that SIF has a good correlation with GPP at the leaf scale, canopy scale, and ecosystem scale. When using SIF satellite remote sensing data to calculate global and regional GPP in existing studies, SIF-GPP models mainly based on linear relationships are often adopted. However, existing studies have shown that there are differences in the SIF-GPP relationships of different plants. When estimating GPP using SIF at the global scale, the SIF-GPP estimation method represented by the linear model will have errors due to the extremely complex vegetation distribution globally, and it is difficult to fully utilize the advantages of SIF from a mechanistic perspective in estimating GPP.

[0006] From the perspective of the structure of the plant photosynthetic system and the mechanism of photosynthesis, most common plants can be divided into C3 plants and C4 plants. Recently, researchers proposed the Mechanistic Light Response (MLR) model, which explicitly considers the key mechanisms connecting chlorophyll fluorescence emission with photosynthesis in C3 and C4 plants. While inheriting the advantages of high theoreticalization, high interpretability, and high generality of the enzyme kinetics model, this theory simplifies some difficult-to-obtain vegetation parameters required for estimating GPP by the enzyme kinetics model by revealing the relationship between SIF and the electron transport rate, providing a new way for estimating GPP. This theory has received extensive attention since its proposal, and recent related studies have proven that there is good consistency between its estimation results and observed values. The MLR model has gradually demonstrated its potential in estimating GPP globally, but there is still no global-scale GPP estimation model constructed based on the MLR theory and remote sensing data, and there is still a large gap in the global application of the MLR theory. Summary of the Invention

[0007] Object of the Invention: The technical problem to be solved by the present invention is to provide a method for estimating the productivity of global land and sea ecosystems based on SIF (sun-induced Chl fluorescence) and light response in view of the deficiencies of the prior art, including the following steps:

[0008] Step 1, preprocess the input data;

[0009] Step 2, calculate key parameters;

[0010] Step 3, calculate the gross primary productivity.

[0011] Step 1 includes: input remote sensing data, including the following categories: temperature data, relative humidity data, fluorescence data, atmospheric carbon dioxide concentration data, vegetation type distribution data, and vegetation physiological data;

[0012] Perform data fusion on the input remote sensing data (Kriging interpolation can generally be used). For the areas with missing data after fusion, use Kriging interpolation to complete them. The formula is:

[0013] ,

[0014] where x 0 is the position where interpolation is currently required; is the interpolation result at the position of x 0 ; n is the number of known pixel points used for interpolation, and the pixel point is a pixel in a remote sensing image; λ i is the Kriging weight for weighting; Z(x i ) is the observed value of the known i-th data point x i ; Check the input data to check whether the spatial range, spatial resolution, temporal resolution, and data type of each type of input data are consistent. If so, it means that the preprocessing of the input data has been completed;

[0015] Otherwise, if the temporal resolution and spatial resolution are inconsistent, downsample the spatial resolution and temporal resolution according to the lowest value of all input data (divide the high-resolution data into multiple blocks, and each block corresponds to a pixel in the low-resolution grid. For each low-resolution pixel, calculate the average value of all high-resolution pixels it covers) to achieve unified temporal and spatial resolution;

[0016] If the data types are inconsistent, unify the input remote sensing data according to the type with the widest representation range (such as selecting the decimal type from the integer type and the decimal type) to ensure compatibility.

[0017] For the case where the spatial ranges are inconsistent, take the input data with the smallest spatial range as the simulation range and perform a cropping operation on the remaining data. For global simulation, the spatial range is usually determined and this problem usually does not need to be considered.

[0018] Step 2 includes: at each pixel point, substitute the input remote sensing data into the mechanistic light response (MLR) theory formula to estimate the Gross Primary Productivity (GPP), and calculate each parameter. Each parameter includes the partial pressure of carbon dioxide CO 2 in the chloroplast stroma C c , the CO 2 compensation point of the chloroplast , the open ratio of the reaction center of the photosynthetic system (The photosynthetic system of plants is divided into two parts, which are academically referred to as photosystem I and photosystem II. The main function of photosystem II is to use light energy to decompose water molecules into oxygen, protons and electrons. This process is called photolysis. Photosystem I receives the electrons transferred from photosystem II and uses light energy to raise the energy of these electrons to a higher level; the reaction center here refers to the reaction center of photosystem II), the energy ratio k of plant heat dissipation and plant fluorescence emission DF , and the ratio of the observed solar-induced chlorophyll fluorescence to the total fluorescence emission, i.e., the escape rate ε.

[0019] In step 2, set the CO 2 partial pressure C c in the chloroplast stroma to be the same as the CO 2 partial pressure C i in the intercellular space. The calculation formula for C i is as follows:

[0020] ,

[0021] where VPD is the saturation vapor pressure deficit and Ψ is the marginal water cost of plant carbon gain; is the partial pressure of carbon dioxide in the atmosphere.

[0022] In step 2, C i can also be calculated by the following formula:

[0023] .

[0024] In step 2, is calculated from the global atmospheric carbon dioxide concentration data:

[0025] ,

[0026] where is the carbon dioxide concentration, taken from the atmospheric carbon dioxide concentration data;

[0027] The CO 2 compensation point of the chloroplast is calculated using the following formula :

[0028] ,

[0029] where is the air temperature, taken from the remote sensing temperature data;

[0030] The saturation vapor pressure deficit VPD is calculated from the relative humidity and air temperature:

[0031] ,

[0032] ,

[0033] where is the saturated water vapor pressure, RH is the relative humidity, taken from remote sensing relative humidity data; exp is the natural exponential function;

[0034] The open ratio is calculated by the following formula :

[0035] ,

[0036] ,

[0037] where SIF is the solar-induced chlorophyll fluorescence data, taken from remote sensing fluorescence data; m is a dimensionless parameter to be fitted, is a constant value of 1 in units of μmol -1 m² s; T opt is the optimal growth temperature of the plant; m opt represents the value of m at the optimal growth temperature T opt of the plant; H d is the decline rate of the peak function above the optimal growth temperature T opt of the plant; is the rise rate of the peak function below the optimal growth temperature T opt of the plant; R is the universal gas constant.

[0038] In step 2, for remote sensing satellite products, the escape rate ε is calculated by the following formula:

[0039] ,

[0040] where NIRv is the near-infrared reflectance of vegetation; is the photosynthetically active radiation component, and the data is taken from vegetation physiological data.

[0041] Step 3 includes: obtaining the distributions of C3 plants and C4 plants using vegetation type distribution data;

[0042] For C3 plants, the formula for calculating GPP is:

[0043] ,

[0044] where is the maximum photosynthetic rate of the photosynthetic system (referring to photosystem II); k DF is the ratio of the energy of plant heat dissipation to the energy of plant-emitted fluorescence;

[0045] For C4 plants, the formula for calculating GPP is:

[0046] ,

[0047] Among them, x is the proportion of the total plant electron transport allocated to the mesophyll tissue.

[0048] The present invention also provides an electronic device, including a processor and a memory, where the memory stores program code, and when the program code is executed by the processor, the processor is caused to execute the steps of the method.

[0049] The present invention also provides a storage medium storing a computer program or instruction, and when the computer program or instruction runs on a computer, it executes the steps of the method.

[0050] This method can be applied to the following fields:

[0051] Global ecosystem monitoring: In the field of ecological protection, this technology is used to monitor the health status and growth dynamics of the global ecosystem, providing a scientific basis for environmental management and protection and supporting sustainable resource utilization.

[0052] Environmental monitoring and protection: The chlorophyll fluorescence inversion technology can be used to detect the response of vegetation to environmental changes, such as climate change, pollutant impacts, etc., thus providing important data support for environmental protection and ecological restoration.

[0053] Global carbon cycle research: By estimating the gross primary productivity of vegetation, this technology provides a new method for studying the changes and distribution of global carbon reserves, supporting the achievement of the goals of carbon peak and carbon neutrality.

[0054] Ecosystem function research: In ecological research, understanding vegetation productivity and photosynthetic efficiency is crucial for studying ecosystem functions and dynamic changes. The chlorophyll fluorescence inversion technology provides an efficient means to obtain this information.

[0055] Beneficial effects: Compared with the current SIF-GPP model mainly based on linear relationships commonly used for inversion, the new scheme proposed by the present invention can more accurately achieve the inversion of global gross primary productivity. The versatility of this scheme has been significantly improved compared with previous schemes and can be directly applied to various types of vegetation. Different from traditional methods such as enzyme kinetics models that require a large number of difficult-to-obtain vegetation physiological parameters during inversion, this scheme greatly simplifies the required number of vegetation parameters during inversion, while ensuring the inversion accuracy, significantly reducing the difficulty of parameter acquisition, and also greatly reducing the result uncertainty caused by parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is the flowchart of the method of the present invention.

[0057] Figure 2 is a schematic diagram of the land surface GPP results simulated by the BEPS model.

[0058] Figure 3 It is a schematic diagram of the land surface GPP result simulated by the NUIST - SIFRM method.

[0059] Figure 4 It is a schematic diagram of the distribution of Pearson correlation coefficients between the GPP estimated by NUIST - SIFRM and the GPP estimated by the model in each continent and ocean.

[0060] Figure 5 It is a schematic diagram of the verification result of the GPP estimated by NUIST - SIFRM using FLUXNET - GPP. Specific implementation manner

[0061] The following further specifically describes the present invention in conjunction with the accompanying drawings and specific implementation manners, and the above - mentioned and / or other advantages of the present invention will become clearer.

[0062] An embodiment of the present invention provides a method for estimating the productivity of global land - sea ecosystems based on SIF and light response, including the following steps:

[0063] Step 1, pre - process the input data;

[0064] The remote sensing data required for the model method constructed by this method includes temperature data, relative humidity data, fluorescence data, atmospheric carbon dioxide concentration data, vegetation type distribution data, and vegetation physiological data (including the near - infrared reflectance NIRv of vegetation and the photosynthetically active radiation component fPAR). Generally speaking, sea and land data are often released separately. Therefore, the present invention first performs data fusion on the sea - land input data, and for the areas with missing data after fusion, Kriging interpolation is used for complementation. Kriging interpolation is a mature and commonly used spatial interpolation method, and the formula is:

[0065] ,

[0066] where x 0 is the position where interpolation is currently required; is the interpolation result at the position of x 0 ; n is the number of known data points used for interpolation; λ i is the Kriging weight for weighting; Z(x i ) is the observed value of the known data point x i ; To ensure the correct execution of the model, the input data is checked to check whether the spatial range, spatial resolution, time resolution, and data type of each type of input data are consistent. After ensuring the consistency of the data, it is considered that the pre - processing of the input data is completed.

[0067] The verification test data used in the present invention includes:

[0068] (1) Global land surface temperature data from 2003 to 2023, and the ERA5 reanalysis land surface temperature data provided by the European Centre for Medium-Range Weather Forecasts is intended to be used;

[0069] (2) Global land surface relative humidity data from 2003 to 2023, and the ERA5 reanalysis relative humidity data provided by the European Centre for Medium-Range Weather Forecasts is intended to be used;

[0070] (3) Global ocean surface temperature data from 2003 to 2023, and the global ocean surface temperature dataset provided by NOAA is intended to be used;

[0071] (4) Global land surface fluorescence data from 2003 to 2023, and the GOSIF product produced by the Global Ecology Group of the University of New Hampshire based on the OCO-2 satellite is intended to be used;

[0072] (5) Global ocean surface fluorescence data from 2003 to 2023, and the ocean fluorescence line height data released by Aqua-MODIS is intended to be used;

[0073] (6) Global atmospheric CO 2 concentration data from 2003 to 2023, and the global atmospheric CO 2 concentration data since 1982 provided by NOAA is intended to be used;

[0074] (7) Land use type data, and the MCD12Q1 IGBP product generated by the MODIS satellite sensor is intended to be used;

[0075] (8) Plant physiological data, and the photosynthetically active radiation component product and vegetation index product provided by the MODIS satellite are intended to be used.

[0076] Step 2, calculate key parameters;

[0077] On the premise of ensuring data consistency, the present invention calculates various parameters based on the input remote sensing data at each pixel point and substitutes them into the MLR theoretical formula for GPP estimation. In the MLR theory, a series of parameters are required for GPP estimation, and the parameters that need to be calculated using the input data include the partial pressure of CO 2 in the chloroplast stroma C c (unit: μbar), the CO 2 compensation point of the chloroplast (unit: μbar); the opening ratio of the reaction center of the photosynthetic system (referring to photosynthetic system II) ; the ratio k of the energy of plant heat dissipation to the energy of plant-emitted fluorescence DF ; the ratio of the observed solar-induced chlorophyll fluorescence to the total emitted fluorescence, i.e., the escape rate ε. The calculation methods of each parameter are as follows:

[0078] In actual calculations, C c can be assumed to be consistent with the partial pressure of intercellular CO 2 (C i ), and C i can be calculated by the following formula:

[0079] ,

[0080] where VPD is the vapor pressure deficit (in hPa), is the partial pressure of carbon dioxide in the atmosphere (in μbar), which can be directly calculated from the global atmospheric CO 2 concentration data:

[0081] ,

[0082] where, is the CO 2 concentration (in ppm). Ψ is the marginal water cost of plant carbon gain (in molH 2 O mol -1 C), usually set to 900. can be described as a function of air temperature:

[0083] ,

[0084] where, is the air temperature (in °C). VPD can be calculated from relative humidity and air temperature:

[0085] ,

[0086] ,

[0087] where, is the saturation water vapor pressure (in hPa), and RH is the relative humidity (in %). For simulation results with a relatively coarse spatial resolution, C i can also be quickly estimated using the following method:

[0088] ,

[0089] can be estimated using the following semi-empirical formula:

[0090] ,

[0091] ,

[0092] where m is a dimensionless parameter to be fitted, In μmol -1 m² s is a constant value of 1; T air is the temperature (in K); T opt is the optimal growth temperature of the plant (in K); m opt Representative in T opt The value of m when H d It is in T opt The rate of decline of the above peak function; It is in T opt The rate of rise of the following peak function; R is the universal gas constant.

[0093] For remote sensing satellite products, ε can be estimated by the following formula:

[0094] ,

[0095] Among them, NIRv is the near-infrared reflectivity of vegetation; is the photosynthetically active radiation component. So far, all intermediate parameters have been calculated. The calculation relationship between input data and intermediate parameters is as follows: Figure 1 shown.

[0096] Step 3, calculate the total primary productivity;

[0097] After the necessary parameters are calculated, for C3 plants, the GPP calculation formula is:

[0098] ,

[0099] in, It is usually set to 0.83; k DF It is the energy ratio of plant heat dissipation and plant fluorescence emission, which is usually set to 19. For C4 plants, the calculation formula of GPP is:

[0100] ,

[0101] Here, x is the proportion of the total electron transport in the plant allocated to the mesophyll tissue, which is usually set to 0.4.

[0102] Model result verification: In order to verify the simulation effect of the present invention, the site observation data set FLUXNET-GPP, the land surface model simulation data set BEPS-GPP and the global ocean GPP product shared by Oregon State University were selected as the verification data of the model. The land surface GPP results simulated by the BEPS model and the land surface GPP results simulated by the NUIST-SIFRM method are shown in Figure 2. Figure 2 and Figure 3 Shown (unit is (10 15), grams of carbon), and the simulation results of the total primary productivity values and trends for each land continent are basically consistent. Figure 4 Shows the distribution of Pearson correlation coefficients between GPP estimated by NUIST-SIFRM and GPP estimated by the model in each continent and the ocean. In each region, especially in the northern hemisphere land area, NUIST-SIFRM and existing GPP products show significant consistency. The verification results of GPP estimated by NUIST-SIFRM using FLUXNET-GPP are as Figure 5 shown. The vertical axis is the result of FLUXNET-GPP, and the horizontal axis is the result of GPP estimated by NUIST-SIFRM. The unit of both is grams of carbon per square meter per day. The global overall correlation coefficient r = 0.7 for both, and the set probability is less than 0.001, proving that the simulation results are highly consistent with the field observation results.

[0103] The present invention provides a method for estimating the productivity of global land and sea ecosystems based on SIF and light response. There are many methods and ways to specifically implement this technical solution. The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Each component not clearly defined in this embodiment can be implemented by existing technologies.

Claims

1. A method for estimating global terrestrial and marine ecosystem productivity based on SIF and light response, characterized in that: The following steps are involved: Step 1, preprocess the input data; Step 2, calculate key parameters; Step 3, calculate the total primary productivity; Step 1 includes: inputting remote sensing data, including the following categories: temperature data, relative humidity data, fluorescence data, atmospheric carbon dioxide concentration data, vegetation type distribution data, and vegetation physiological data; Data fusion is performed on the input remote sensing data. For the missing areas after fusion, Kriging interpolation is used to fill them. The formula is: Among them, x0 is the current position that needs to be interpolated; Z * (x0) is the interpolation result at x0; n is the number of known pixel points used for interpolation, and a pixel point is a pixel point in a remote sensing image; λ i is the kriging weight used for weighting; Z(x i ) is the known i-th data point x i Observation values; Check the input data to see whether the spatial range, spatial resolution, temporal resolution, and data type of each type of input data are consistent. If yes, it means that the input data preprocessing has been completed; Otherwise, if the temporal resolution and spatial resolution are inconsistent, the spatial resolution and temporal resolution are downsampled according to the lowest value of all input data to achieve uniform temporal and spatial resolutions; If the data types are inconsistent, the input remote sensing data will be unified according to the type with the widest representation range; Step 2 includes: at each pixel point, the input remote sensing data is brought into the mechanical light response MLR theoretical formula to estimate the total primary productivity GPP, and various parameters are calculated, including the partial pressure of carbon dioxide CO2 in the chloroplast matrix C c , CO2 compensation point of chloroplasts Γ * , the open ratio q of the photosynthetic system reaction center L , the energy ratio k of plant heat dissipation and plant fluorescence emission DF , and the ratio of the observed solar-induced chlorophyll fluorescence to the total emitted fluorescence, i.e., the escape velocity ε; Step 3 includes: obtaining the distribution of C3 plants and C4 plants using vegetation type distribution data; For C3 plants, the calculation formula for gross primary productivity GPP is: Among them, Φ PSIImax is the maximum photosynthetic rate of the photosynthetic system; k DF is the energy ratio of plant heat dissipation and plant fluorescence emission; For C4 plants, the calculation formula for gross primary productivity GPP is: Where x is the proportion of the total electron transport in the plant allocated to the mesophyll tissue; In step 2, C is calculated from the global atmospheric carbon dioxide concentration data. a : in, is the carbon dioxide concentration; The CO2 compensation point Γ of chloroplasts is calculated using the following formula: * : Γ * =36.9+1.18×(T air -25)+0.036×(T air -25) 2 , Among them, T air is the air temperature, obtained from remote sensing temperature data; Calculate the saturated water vapor pressure difference VPD by relative humidity and air temperature: VPD=(1-RH)×e s , Among them, e s is the saturated water vapor pressure, RH is the relative humidity; exp is the natural exponential function; The open ratio q is calculated by the following formula L : Where SIF is the solar induced chlorophyll fluorescence data; m is a dimensionless parameter to be fitted, and δ is in μmol. -1 m 2 s is a constant value of unit; T opt is the optimal growth temperature of plants; m opt Represents the optimal growth temperature of the plant opt The value of m when H d The optimal growth temperature of the plant is T opt The rate of decrease of the peak function above; H a The optimal growth temperature of the plant is T opt The rate of rise of the following peak function; R is the universal gas constant; In step 2, for remote sensing satellite products, the escape velocity ε is calculated using the following formula: Among them, NIRv is the near-infrared reflectivity of vegetation; f PAR is the photosynthetically active radiation component; In step 2, the CO2 partial pressure in the chloroplast matrix is ​​set to C c The partial pressure of CO2 between cells i Same, C i The calculation formula is: Where VPD is the saturated water vapor pressure difference, Ψ is the marginal water cost of plant carbon gain, C a is the partial pressure of carbon dioxide in the atmosphere; Or calculate C by the following formula i : C i =0.7×C a 。 2. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores program codes, and when the program codes are executed by the processor, the processor executes the steps of the method according to claim 1.

3. A storage medium, characterized in that: A computer program or instruction is stored, and when the computer program or instruction is run on a computer, the steps of the method according to claim 1 are executed.

Images