Carbon-water key element collaborative inversion method and system based on data driving

Abstract

The invention relates to the technical field of carbon-water flux remote sensing inversion, and provides a carbon-water key element collaborative inversion method and system based on data driving, and the method comprises the steps: obtaining data needed by the carbon-water key element collaborative inversion based on data driving, and constructing a carbon-water key element collaborative inversion database; preprocessing the carbon-water key element collaborative inversion database and obtaining a carbon-water key element collaborative inversion driving data set; constructing a carbon-water key element collaborative inversion model; and training the carbon-water key element collaborative inversion model to obtain an optimal collaborative inversion model, and performing collaborative inversion on global evapotranspiration, total primary productivity and ecological system water utilization efficiency, thereby solving the problems that an existing data-driven model cannot effectively consider a coupling relationship between the total primary productivity and evapotranspiration, the result is not accurate, and the reliability is poor. The difference between the data and observation data is large, and large uncertainty is brought to the change and attribution research of the water utilization efficiency of the global ecosystem.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon-water flux remote sensing inversion, and in particular to a data-driven carbon-water key element collaborative inversion method and system. Background Art

[0002] The key carbon-water elements include total primary productivity, evapotranspiration and ecosystem water use efficiency, and their accurate estimation is crucial for water resources management and food security. Among them, ecosystem water use efficiency is the ratio of total primary productivity to evapotranspiration, which connects the carbon-water cycle. At the global and regional scales, the remote sensing inversion methods of evapotranspiration and total primary productivity are quite mature, while ecosystem water use efficiency is mainly obtained by estimating the total primary productivity and evapotranspiration at the global and regional scales. The existing total primary productivity, evapotranspiration and ecosystem water use efficiency estimation models include two categories, non-carbon-water coupled products and carbon-water coupled products. Non-carbon-water coupled products, such as MODIS products, estimate total primary productivity and evapotranspiration separately, and then solve for ecosystem water use efficiency; carbon-water coupled products, such as PML_V2 products, use the stomatal conductance model to couple vegetation photosynthesis and transpiration, and simultaneously estimate evapotranspiration and total primary productivity, and then solve for ecosystem water use efficiency. Although existing models can obtain accurate and reliable evapotranspiration and gross primary productivity, the results of ecosystem water use efficiency calculated from the two are less accurate. The inversion errors of gross primary productivity and evapotranspiration will increase the errors and uncertainties in the estimation results of ecosystem water use efficiency.

[0003] With the emergence of data-driven models such as machine learning and deep learning, although the estimation accuracy of gross primary productivity and evapotranspiration has been improved, they are still independently modeled and estimated, and then the ecosystem water use efficiency is obtained by calculating the ratio. The coupling relationship between gross primary productivity and evapotranspiration has not been effectively considered, making the obtained ecosystem water use efficiency results inaccurate and significantly different from the observed results, which has brought great uncertainty to the changes and attribution studies of global ecosystem water use efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a data-driven carbon-water key element collaborative inversion method and system, aiming to solve the problem that the existing data-driven models fail to effectively consider the coupling relationship between total primary productivity and evapotranspiration, resulting in inaccurate results of ecosystem water use efficiency, which are significantly different from the observed results, and bring great uncertainty to the changes and attribution studies of global ecosystem water use efficiency.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] A data-driven carbon-water key element collaborative inversion method includes the following steps:

[0007] Obtain the data required for data-driven carbon-water key element collaborative inversion and build a corresponding carbon-water key element collaborative inversion database; wherein the carbon-water key elements include ecosystem water use efficiency, gross primary productivity and evapotranspiration;

[0008] Preprocess the data in the carbon-water key element collaborative inversion database and obtain the collaborative inversion driving data set;

[0009] Construct a data-driven carbon-water key element collaborative inversion model, using the collaborative inversion driving dataset as the input of the carbon-water key element collaborative inversion model;

[0010] Based on the collaborative inversion driving data set, the data-driven carbon-water key element collaborative inversion model is trained. The cost function is set during training, and the model parameters with the minimum cost function are obtained through the grid search method to obtain the optimal data-driven carbon-water key element collaborative inversion model;

[0011] Through the optimal data-driven carbon-water key element collaborative inversion model, based on the corresponding remote sensing products and meteorological reanalysis products, the global ecosystem water use efficiency, total primary productivity and evapotranspiration are collaboratively inverted.

[0012] Optionally, the data required for the data-driven collaborative inversion of carbon-water key elements include: global flux network site observation data, remote sensing products and meteorological reanalysis products.

[0013] Optionally, the specific process of preprocessing the carbon-water key element collaborative inversion database and obtaining the collaborative inversion driving data set is:

[0014] Based on the total primary productivity and evapotranspiration observed at the global flux network site, the ecosystem water use efficiency observed at the site is obtained. The expression of the ecosystem water use efficiency observed at the site is shown in the following formula (1):

[0015]

[0016] Among them, eWUE obs represents the ecosystem water use efficiency calculated from site observations, GPP obs represents the total primary productivity observed at the site, ET obs represents the evapotranspiration observed at the site;

[0017] According to the longitude and latitude of the global flux network stations, the data values ​​of the meteorological reanalysis grids and remote sensing pixels corresponding to each global flux network station are extracted, and the corresponding meteorological reanalysis grids and remote sensing pixels are combined with the observation data of the global flux network stations to obtain the collaborative inversion driven dataset.

[0018] Optionally, the total primary productivity is obtained directly through global flux network site observation data or through separation and solution of net ecosystem carbon exchange data in global flux network site observation data.

[0019] Optionally, the data-driven carbon-water key element collaborative inversion model is:

[0020] The expression of the data-driven carbon-water key element collaborative inversion model is as follows (2):

[0021] ∪(eWUE, GPP, ET)=

[0022] f(WS, RH, VPD, T a ,P,R n , R s , R l ,SIF,LST,NDyI,FPAR,FVC,LAI,α,ele,T s , SM) (2)

[0024] Among them, eWUE is the ecosystem water use efficiency, GPP is the gross primary productivity, ET is the evapotranspiration, WS is the wind speed, RH is the relative humidity, VPD is the saturated vapor pressure deficit, T a is the air temperature, P is the atmospheric pressure, R n is the net radiation, R s is the incident shortwave radiation, R l is the incident longwave radiation, SIF is chlorophyll fluorescence, LST is the land surface temperature, NDyI is the normalized difference vegetation index, FPAR is the ratio of light and effective radiation absorption, FVC is the vegetation cover, LAI is the leaf area index, α is the albedo, ete is the elevation, T s is soil temperature, and SM is soil moisture content.

[0025] Optionally, the specific process of using the collaborative inversion driving data set as the input of the data-driven carbon-water key element collaborative inversion model is:

[0026] All available data in the collaborative inversion driven dataset were randomly sampled and divided into 80% model training set and 20% model validation set according to each vegetation type. For each vegetation type, the data-driven carbon-water key element collaborative inversion model was trained by the 80% model training set using the ten-fold cross-validation method, and the ecological water use efficiency inversion model was validated by the 20% model validation set.

[0027] Optionally, the cost function is:

[0028] The expression of the cost function is shown in the following formula (3):

[0029]

[0030] Among them, cost(∪(eWUE, GPP, ET)) represents the cost function, eWUE est is the estimated value of ecosystem water use efficiency, eWUE obs is the ecosystem water use efficiency calculated from site observations, a is the coefficient of the ecosystem water use efficiency term, GPP est is an estimate of gross primary productivity, GPP obs is the total primary productivity observed at the site or the total primary productivity obtained by separating the net ecosystem carbon exchange observed at the site, b is the coefficient to be determined for the total primary productivity term, ET est is the estimated value of evapotranspiration, ET obs is the evapotranspiration observed at the site, and c is the coefficient to be determined for the evapotranspiration term.

[0031] The present invention also provides a data-driven carbon-water key element collaborative inversion system, comprising:

[0032] Data acquisition module, used to obtain global flux site observation data, remote sensing products and meteorological reanalysis products;

[0033] Data preprocessing module, used to preprocess the collected global flux network site observation data, remote sensing products and meteorological reanalysis products;

[0034] Model building and training module, used to build and train a data-driven carbon-water key element collaborative inversion model;

[0035] The inversion calculation module is used to perform the co-inversion of ecosystem water use efficiency, total primary productivity and evapotranspiration based on the corresponding global flux site observation data, remote sensing products and meteorological reanalysis products using the optimal data-driven carbon-water key element co-inversion model;

[0036] The result verification and correction module is used to verify and correct the data set using the carbon-water key element co-inversion drive.

[0037] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-mentioned data-driven carbon-water key element collaborative inversion method.

[0038] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the above-mentioned data-driven carbon-water key element collaborative inversion method is implemented.

[0039] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0040] The present invention provides a data-driven carbon-water key element collaborative inversion method, comprising the following steps: obtaining data required for data-driven carbon-water key element collaborative inversion, and constructing a corresponding carbon-water key element collaborative inversion database; wherein the carbon-water key elements include ecosystem water use efficiency, total primary productivity and evapotranspiration; preprocessing the data in the carbon-water key element collaborative inversion database and obtaining a collaborative inversion driving data set; constructing a data-driven carbon-water key element collaborative inversion model, using the collaborative inversion driving data set as the input of the carbon-water key element collaborative inversion model; training the data-driven carbon-water key element collaborative inversion model based on the collaborative inversion driving data set, setting a cost function during training, and obtaining the cost function through a grid search method. The model parameters of the function are minimized and the optimal data-driven carbon-water key element collaborative inversion model is obtained; through the optimal data-driven carbon-water key element collaborative inversion model, according to the corresponding remote sensing products and meteorological reanalysis products, the global ecosystem water use efficiency, total primary productivity and evapotranspiration are collaboratively inverted; it not only ensures that evapotranspiration, total primary productivity and ecosystem water use efficiency have high inversion accuracy, but also solves the problem of inaccurate ecosystem water use efficiency estimation results caused by the existing data-driven model independently modeling evapotranspiration and total primary productivity and failing to effectively consider the coupling relationship between total primary productivity and evapotranspiration, thereby further improving the reliability of the global total primary productivity, evapotranspiration and ecosystem water use efficiency change and attribution research. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic flow chart of the data-driven carbon-water key element collaborative inversion method in Example 1 of the present invention;

[0042] Figure 2It is a structural schematic diagram of the data-driven carbon-water key element collaborative inversion system in the second embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0044] Example 1

[0045] Reference Figure 1 , a data-driven carbon-water key element collaborative inversion method, comprising the following steps:

[0046] Step 1: Obtain the data required for data-driven carbon-water key element collaborative inversion and construct a corresponding carbon-water key element collaborative inversion database; wherein the carbon-water key elements include ecosystem water use efficiency, gross primary productivity and evapotranspiration;

[0047] In this embodiment, the data required for the data-driven coordinated inversion of carbon-water key elements include: global flux network site observation data, remote sensing products, and meteorological reanalysis products;

[0048] The global flux network sites include: Fluxnet, Ameriflux, OzFlux, EuroFlux, Heihe Ecohydrological Remote Sensing Experiment HiWATER, Lanzhou University Cold and Arid Regions Scientific Observation Network CARN and other international and domestic flux observation network sites;

[0049] The observation data of the global flux network stations include: station longitude and latitude data, automatic weather station observation data, eddy correlator observation data and ground vegetation height measurement data;

[0050] The observation data of the automatic weather station include: wind speed, air humidity, soil moisture, air temperature, soil temperature, atmospheric pressure, vapor pressure deficit, net radiation, incident shortwave radiation and incident longwave radiation;

[0051] Eddy correlator observation data include: friction velocity, evapotranspiration, gross primary productivity and net ecosystem carbon exchange;

[0052] Remote sensing products include: GOSIF chlorophyll fluorescence, MOD11A1 surface temperature, MCD43A4 reflectivity, MCD15A3H photosynthetically active radiation absorption ratio, GLASS vegetation cover, GLASS leaf area index, GLASS albedo and GMTED2010 elevation products;

[0053] Meteorological reanalysis products include: wind speed, relative humidity, saturated water vapor pressure deficit, air temperature, atmospheric pressure, soil moisture content, net radiation, incident shortwave radiation and incident longwave radiation;

[0054] Meteorological reanalysis products may be ERA5-Land reanalysis products, ERA5 reanalysis products, ERA-Interim reanalysis products, MERRA-2 reanalysis products or JRA55 reanalysis products, etc.

[0055] Step 2: preprocess the carbon-water key element collaborative inversion database and obtain the carbon-water key element collaborative inversion driving data set;

[0056] In this embodiment, the specific process of preprocessing the carbon-water key element collaborative inversion database and obtaining the carbon-water key element collaborative inversion driving data set is:

[0057] Based on the total primary productivity and evapotranspiration observed at the global flux network sites, the ecosystem water use efficiency calculated from the site observations was obtained. The expression of the ecosystem water use efficiency calculated from the site observations is shown in the following formula (1):

[0058]

[0059] Among them, eWUE obs represents the ecosystem water use efficiency calculated from site observations, GPP obs represents the total primary productivity observed at the site, ET obs represents the evapotranspiration observed at the site;

[0060] In this embodiment, the MODIS normalized vegetation index is calculated from the MCD43A4 reflectance product;

[0061] According to the longitude and latitude of the global flux network stations, the data values ​​of the meteorological reanalysis grids and remote sensing pixels corresponding to each global flux network station are extracted. The obtained data values ​​of the corresponding meteorological reanalysis grids and remote sensing pixels are combined with the observation data of the global flux network stations to obtain the carbon-water key element collaborative inversion driving data set.

[0062] The total primary productivity is directly obtained through the observation data of the global flux network sites or is obtained by separating and solving the net ecosystem carbon exchange data in the observation data of the global flux network sites; in this embodiment, if the total primary productivity is not directly provided in the observation data of some global flux network sites, but the net ecosystem carbon exchange data is provided, the total primary productivity is separated and solved from the net ecosystem carbon exchange by a corresponding method; in this embodiment, the corresponding method is a daytime-based method or a nighttime-based method. If the total primary productivity of some sites is separated and solved by the net ecosystem carbon exchange, when it is combined with other sites that provide total primary productivity, it should be noted that the total primary productivity separated and solved by the daytime-based method of the site that does not provide the total primary productivity should be combined with the total primary productivity of the daytime-based method of the site that provides the total primary productivity; or the total primary productivity separated and solved by the nighttime-based method of the site that does not provide the total primary productivity should be combined with the total primary productivity of the nighttime-based method of the site that provides the total primary productivity.

[0063] Step 3: Construct a data-driven carbon-water key element collaborative inversion model, using the collaborative inversion driving dataset as the input of the carbon-water key element collaborative inversion model, and using the evapotranspiration, gross primary productivity and ecosystem water use efficiency observed by the coupled global flux network sites as the output of the data-driven carbon-water key element collaborative inversion model;

[0064] In this embodiment, the data-driven carbon-water key element collaborative inversion model is a deep learning model, and the data-driven carbon-water key element collaborative inversion model is:

[0065] The expression of the data-driven carbon-water key element collaborative inversion model is as follows (2):

[0066] ∪(eWUE, GPP, ET)

[0067] =

[0068] f(WS, RH, VPD, T a ,P,R n , R s , R l ,SIF,LST,NDyI,FPAR,FVC,LAI,α,ele,T s , SM)(2)

[0069] Among them, eWUE is the ecosystem water use efficiency, GPP is the gross primary productivity, ET is the evapotranspiration, WS is the wind speed, RH is the relative humidity, VPD is the saturated vapor pressure deficit, T a is the air temperature, P is the atmospheric pressure, R n is the net radiation, R s is the incident shortwave radiation, R l is the incident longwave radiation, SIF is chlorophyll fluorescence, LST is the land surface temperature, NDyI is the normalized difference vegetation index, FPAR is the ratio of light and effective radiation absorption, FVC is the vegetation cover, LAI is the leaf area index, α is the albedo, ele is the elevation, T s is soil temperature, and SM is soil moisture content.

[0070] In this embodiment, the specific process of using the collaborative inversion driving data set as the input of the carbon-water key element collaborative inversion model is as follows:

[0071] All available data in the collaborative inversion driven dataset were randomly sampled and divided into 80% model training set and 20% model validation set according to each vegetation type. For each vegetation type, the data-driven carbon-water key element collaborative inversion model was trained by the 80% model training set using the ten-fold cross-validation method, and the ecosystem water use efficiency inversion model was validated by the 20% model validation set.

[0072] Step 4: training the data-driven carbon-water key element collaborative inversion model based on the collaborative inversion driving data set, setting a cost function during training, obtaining the model parameters with the minimum cost function through a grid search method, and obtaining the optimal data-driven carbon-water key element collaborative inversion model;

[0073] In this embodiment, the cost function is:

[0074] The expression of the cost function is shown in the following formula (3):

[0075]

[0076] Among them, cost(∪(eWUE, GPP, ET)) represents the cost function, eWUE est is the estimated value of ecosystem water use efficiency, eWUE obs is the ecosystem water use efficiency calculated from site observations, a is the coefficient of the ecosystem water use efficiency term, GPP est is an estimate of gross primary productivity, GPP obs is the total primary productivity observed at the site or the total primary productivity obtained by separating the net ecosystem carbon exchange observed at the site, b is the coefficient to be determined for the total primary productivity term, ETest is the estimated value of evapotranspiration, ET obs is the evapotranspiration observed at the site, and c is the coefficient to be determined for the evapotranspiration term.

[0077] Step 5. Through the optimal data-driven carbon-water key element collaborative inversion model, according to the corresponding remote sensing products and meteorological reanalysis products, collaboratively invert the global ecosystem water use efficiency, gross primary productivity and evapotranspiration.

[0078] Example 2

[0079] In order to execute the method corresponding to the above embodiment 1 to achieve the corresponding functions and technical effects, refer to Figure 2 In this embodiment, a data-driven carbon-water key element collaborative inversion system is provided, including:

[0080] The data acquisition module 101 is used to obtain the observation data, remote sensing products and meteorological reanalysis products of the global flux network sites, including the latitude and longitude data of the global flux network sites such as Fluxnet, Ameriflux, OzFlux, EuroFlux, Heihe Ecohydrological Remote Sensing Experiment HiWATER, Lanzhou University Cold and Arid Region Scientific Observation Network CARN, automatic weather station observation data, eddy correlator observation data and ground vegetation height measurement data; remote sensing products such as GOSIF chlorophyll fluorescence, MOD11A1 surface temperature, MCD43A4 reflectivity, MCD15A3H photosynthetically active radiation absorption ratio, GLASS vegetation cover, GLASS leaf area index, GLASS albedo and GMTED2010 elevation; meteorological reanalysis products such as ERA5-Land reanalysis products, ERA5 reanalysis products, ERA-Interim reanalysis products, MERRA-2 reanalysis products or JRA55 reanalysis products.

[0081] The data preprocessing module 102 is used to perform data preprocessing on the collected global flux network site observation data, remote sensing products and meteorological reanalysis products, including: obtaining the ecosystem water use efficiency calculated from the site observation based on the total primary productivity and evapotranspiration observed by the global flux network site; if the total primary productivity is not directly provided in the observation data of some global flux network sites, but the net ecosystem carbon exchange data is provided, the total primary productivity is obtained by separating and solving from the net ecosystem carbon exchange through the daytime-based method or the nighttime-based method; according to the longitude and latitude of the global flux network site, the data values ​​of the meteorological reanalysis grid and remote sensing pixel corresponding to each global flux network site are extracted, and the corresponding meteorological reanalysis grid and remote sensing pixel data values ​​obtained are combined with the observation data of the global flux network site. Through data preprocessing, a comprehensive and accurate data foundation with consistent format and resolution is provided for subsequent model construction and inversion calculation.

[0082] The model building and training module 103 is used to establish and train a data-driven carbon-water key element collaborative inversion model; based on a data-driven carbon-water key element collaborative inversion method in Example 1, a data-driven carbon-water key element collaborative inversion model is constructed and the model is trained accordingly to obtain the optimal data-driven carbon-water key element collaborative inversion model.

[0083] The inversion calculation module 104 is used to perform the collaborative inversion of ecosystem water use efficiency, total primary productivity and evapotranspiration based on the corresponding remote sensing products and meteorological reanalysis products using the optimal data-driven carbon-water key element collaborative inversion model.

[0084] The result verification and correction module 105 is used to use the carbon-water key element collaborative inversion driving data set for verification and correction, evaluate the accuracy and reliability of the model estimation through comparative analysis, and make necessary adjustments and optimizations to the model.

[0085] The data-driven carbon-water key element collaborative inversion system in this embodiment integrates multiple functional modules such as data collection, data preprocessing, model construction and training, inversion calculation, and result verification and correction, realizing the full automation processing from data to results; using advanced algorithms and models, through the collaborative inversion of evapotranspiration, total primary productivity and ecosystem water use efficiency, evapotranspiration and total primary productivity are coupled, and the estimation accuracy of ecosystem water use efficiency is improved; supporting real-time data processing and result output, providing timely and effective decision support for ecological water resources management and agricultural irrigation optimization; the system architecture is flexible, and functional modules and data sources can be expanded according to actual needs to adapt to the application needs of different regions and scenarios; it is widely used in agriculture, forestry, water resources management and environmental protection and other fields, providing technical support and decision-making basis for the efficient utilization and sustainable development of ecological water resources. For example, in agricultural irrigation, by estimating crop water use efficiency, optimizing irrigation plans, and improving water resource utilization efficiency; in ecological environment monitoring, the use of water by the ecosystem is evaluated to provide a scientific basis for ecological protection and environmental restoration.

[0086] Example 3

[0087] This embodiment provides an electronic device, including a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the data-driven carbon-water key element collaborative inversion method of Example 1.

[0088] Optionally, the above-mentioned electronic device may be a server.

[0089] In addition, this embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the data-driven carbon-water key element collaborative inversion method of Example 1 is implemented.

[0090] It is understandable that the processor in the embodiments of the present invention may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0091] The method steps in the embodiments of the present invention can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0092] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a storage medium or transmitted through the storage medium. The computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0093] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data-driven carbon-water key element collaborative inversion method, characterized in that: The following steps are involved: Obtain the data required for data-driven carbon-water key element collaborative inversion and build a corresponding carbon-water key element collaborative inversion database; wherein the carbon-water key elements include ecosystem water use efficiency, gross primary productivity and evapotranspiration; Preprocess the data in the carbon-water key element collaborative inversion database and obtain the collaborative inversion driving data set; Construct a data-driven carbon-water key element collaborative inversion model, using the collaborative inversion driving dataset as the input of the carbon-water key element collaborative inversion model; Based on the collaborative inversion driving data set, the data-driven carbon-water key element collaborative inversion model is trained. The cost function is set during training, and the model parameters with the minimum cost function are obtained through the grid search method to obtain the optimal data-driven carbon-water key element collaborative inversion model; Through the optimal data-driven carbon-water key element collaborative inversion model, based on the corresponding remote sensing products and meteorological reanalysis products, the global ecosystem water use efficiency, total primary productivity and evapotranspiration are collaboratively inverted.

2. The data-driven carbon-water key element collaborative inversion method according to claim 1, characterized in that: The data required for the data-driven collaborative inversion of carbon-water key elements include: global flux network site observation data, remote sensing products and meteorological reanalysis products.

3. The data-driven carbon-water key element collaborative inversion method according to claim 2, characterized in that: The specific process of preprocessing the carbon-water key element collaborative inversion database and obtaining the collaborative inversion driving data set is as follows: Based on the total primary productivity and evapotranspiration observed at the global flux network site, the ecosystem water use efficiency observed at the site is obtained. The expression of the ecosystem water use efficiency observed at the site is shown in the following formula (1): Among them, eWUE obs represents the ecosystem water use efficiency calculated from site observations, GPP obs represents the total primary productivity observed at the site, ET obs represents the evapotranspiration observed at the site; According to the longitude and latitude of the global flux network stations, the data values ​​of the meteorological reanalysis grids and remote sensing pixels corresponding to each global flux network station are extracted, and the corresponding meteorological reanalysis grids and remote sensing pixels are combined with the observation data of the global flux network stations to obtain the collaborative inversion driven dataset.

4. The data-driven carbon-water key element collaborative inversion method according to claim 3, characterized in that: The total primary productivity is obtained directly through the global flux network site observation data or through separation and solution of the net ecosystem carbon exchange data in the global flux network site observation data.

5. The data-driven carbon-water key element collaborative inversion method according to claim 1, characterized in that: The data-driven carbon-water key element collaborative inversion model is: The expression of the data-driven carbon-water key element collaborative inversion model is as follows (2): ∪(eWUE,GPP,ET)= f(WS,RH,VPD,T a ,P,R n ,R s ,R l ,SIF,LST,NDVI,FPAR,FVC,LAI,α,ele,T s ,SM)(2) Among them, eWUE is the ecosystem water use efficiency, GPP is the gross primary productivity, ET is the evapotranspiration, WS is the wind speed, RH is the relative humidity, VPD is the saturated vapor pressure deficit, T a is the air temperature, P is the atmospheric pressure, R n is the net radiation, R s is the incident shortwave radiation, R l is the incident longwave radiation, SIF is chlorophyll fluorescence, LST is the land surface temperature, NDVI is the normalized difference vegetation index, FPAR is the ratio of light and effective radiation absorption, FVC is the vegetation cover, LAI is the leaf area index, α is the albedo, ele is the elevation, T s is soil temperature, and SM is soil moisture content.

6. The data-driven carbon-water key element collaborative inversion method according to claim 1, characterized in that: The specific process of using the collaborative inversion driving data set as the input of the data-driven carbon-water key element collaborative inversion model is as follows: All available data in the collaborative inversion driven dataset were randomly sampled and divided into 80% model training set and 20% model validation set according to each vegetation type. For each vegetation type, the data-driven carbon-water key element collaborative inversion model was trained by the 80% model training set using the ten-fold cross-validation method, and the data-driven carbon-water key element collaborative inversion model was validated by the 20% model validation set.

7. The data-driven carbon-water key element collaborative inversion method according to claim 1, characterized in that: The cost function is: The expression of the cost function is shown in the following formula (3): Among them, cost(∪(eWUE,GPP,ET)) represents the cost function, eWUE est is the estimated value of ecosystem water use efficiency, eWUE obs is the ecosystem water use efficiency calculated from site observations, a is the coefficient of the ecosystem water use efficiency term, GPP est is an estimate of gross primary productivity, GPP obs is the total primary productivity observed at the site or the total primary productivity obtained by separating the net ecosystem carbon exchange observed at the site, b is the coefficient to be determined for the total primary productivity term, ET est is the estimated value of evapotranspiration, ET obs is the evapotranspiration observed at the site, and c is the coefficient to be determined for the evapotranspiration term.

8. A data-driven carbon-water key element collaborative inversion system, characterized in that: include: Data acquisition module, used to obtain global flux site observation data, remote sensing products and reanalysis products; Data preprocessing module, used to preprocess the collected global flux network site observation data, remote sensing products and meteorological reanalysis products; Model building and training module, used to build and train a data-driven carbon-water key element collaborative inversion model; The inversion calculation module is used to perform the co-inversion of ecosystem water use efficiency, total primary productivity and evapotranspiration based on the corresponding global flux site observation data, remote sensing products and reanalysis products using the optimal data-driven carbon-water key element co-inversion model; The result verification and correction module is used to verify and correct the data set using the carbon-water key element co-inversion drive.

9. An electronic device, characterized in that: It comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the data-driven carbon-water key element collaborative inversion method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the data-driven carbon-water key element collaborative inversion method according to any one of claims 1 to 7 is implemented.

Images