CN114331233A - Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium - Google Patents
Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114331233A CN114331233A CN202210249453.2A CN202210249453A CN114331233A CN 114331233 A CN114331233 A CN 114331233A CN 202210249453 A CN202210249453 A CN 202210249453A CN 114331233 A CN114331233 A CN 114331233A
- Authority
- CN
- China
- Prior art keywords
- vegetation
- remote sensing
- pixel
- sensing image
- total primary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The application provides a method and a device for estimating total primary productivity of vegetation, electronic equipment and a storage medium, which relate to the technical field of remote sensing image processing, and specifically comprise the following steps: acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period; acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image; based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, the characteristic quantity of each pixel of the remote sensing image is calculated, and the method comprises the following steps: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor; and inputting the characteristic quantity of each pixel of the remote sensing image into a pre-trained vegetation total primary productivity estimation model, and outputting a vegetation total primary productivity estimation value of each pixel. The method can improve the precision of the estimation of the total primary productivity of the vegetation.
Description
Technical Field
The application relates to the technical field of remote sensing image processing, in particular to a vegetation total primary productivity estimation method and device, electronic equipment and a storage medium.
Background
The Gross Primary Productivity (GPP) is an important factor for estimating the earth bearing capacity and evaluating the sustainable development of the terrestrial ecosystem, plays an important role in global changes and carbon cycles, and is particularly important for accurate evaluation of the GPP. GPP refers to the fixation of the total amount of organic carbon per unit time by green plants in the ecosystem through the photosynthetic pathway, determining the initial energy and total amount of material entering the terrestrial ecosystem. The international general estimation GPP method mainly comprises methods of flux station continuous observation, land ecological process model estimation and the like. The remote sensing data combined GPP estimation model realizes estimation of vegetation GPP with continuous space and without damage to vegetation. Remote sensing estimation GPP models are mainly classified into 3 types: the system comprises an empirical vegetation index model, a vegetation ecological process model and a machine learning model.
The remote sensing Light energy utilization rate model estimates the net primary productivity on land by using Absorbed Photosynthetically Active Radiation (APAR) and Light Use Efficiency (LUE), considers the influence of environmental stress factors such as nutrients, water, temperature and the like on vegetation photosynthesis, and is a model for producing GPP by combining a vegetation ecological process with more applications. The light energy utilization rate model input data mainly come from remote sensing data, the dependence on-site observation data is reduced or avoided, and the method is widely applied to regional and global-scale GPP estimation research. At present, global parameterization is carried out on some parameters in a model on the basis of a light energy utilization rate model, and a global GPP remote sensing product can be produced by combining long-time sequence data. However, the existing GPP estimation models and algorithms have some disadvantages, and the existing products have different product accuracies in different regions around the world, which is difficult to meet the requirements of practical application.
Disclosure of Invention
In view of the above, the present application provides a method, an apparatus, an electronic device and a storage medium for estimating the total primary productivity of vegetation, which can solve the technical problem of low estimation accuracy of the total primary productivity of vegetation.
In a first aspect, an embodiment of the present application provides a method for estimating total primary productivity of vegetation, including:
acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period;
acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image;
calculating the characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, wherein the characteristic quantity comprises the following components: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
and inputting the characteristic quantity of each pixel of the remote sensing image into a pre-trained vegetation total primary productivity estimation model, and outputting a vegetation total primary productivity estimation value of each pixel.
Further, the step of calculating the photosynthetically active radiation ratio comprises:
obtaining enhanced vegetation index EVI of each pixel of the remote sensing image, and calculating a first photosynthetically active radiation ratio FPAREVI:
Wherein β is a parameter;
obtaining leaf area index of each pixel of remote sensing imageLAICalculating a second photosynthetically active radiation ratio FPARLAI:
Wherein K is the radiation extinction coefficient.
Further, the calculating of the moisture limiting factor comprises:
obtaining land surface moisture index LSWI of each pixel based on SWIR band information and NIR band information of each pixel of the remote sensing image, and calculating a first moisture limiting factorW LSWI:
Wherein the content of the first and second substances,LSWI max is the maximum LSWI value of the vegetation growing season in each pixel;
acquiring saturated water vapor pressure difference of each pixel of remote sensing image𝑉𝑃𝐷Calculating a second moisture limiting factorW VPD:
Wherein the content of the first and second substances,VPD max andVPD min the maximum value and the minimum value of the saturated water vapor pressure difference are determined according to the vegetation type; when in use𝑉𝑃𝐷 > VPD max When the temperature of the water is higher than the set temperature,W VPDis 0, when𝑉𝑃𝐷 <VPD min When the temperature of the water is higher than the set temperature,W VPDis 1.
Further, the step of calculating the temperature limiting factor comprises:
calculating a first temperature limiting factorT TEM :
When in useT min < T < T max The method comprises the following steps:
if not, then,T TEM =0;
in the formula (I), the compound is shown in the specification,Tin order to monitor the average temperature over a period of time,T opt adopting the temperature corresponding to the maximum EVI of each pixel enhanced vegetation index as the temperature corresponding to the maximum photosynthesis rate of vegetation;T min andT max respectively the lowest temperature and the highest temperature when the vegetation is subjected to photosynthesis,T min =0℃,T max =T opt + ( T opt -T min )2;
calculating a second temperature limiting factor using a CASA algorithmT s:
Wherein the content of the first and second substances,T S 1in order to reduce the factors of the productivity of vegetation,T S 2the trend factor shows that the light energy utilization rate of the plant is gradually reduced.
Further, the vegetation total primary productivity estimation model adopts a BP neural network, and the training process of the vegetation total primary productivity estimation model comprises the following steps:
acquiring a training set; the training set comprises a plurality of historical remote sensing image samples, corresponding meteorological observation data samples and flux tower data;
acquiring spatial position information of each pixel of each historical remote sensing image sample, and rasterizing meteorological observation data to obtain meteorological observation data of each pixel of each historical remote sensing image sample; rasterizing the flux tower data to obtain flux tower data of each pixel of each historical remote sensing image sample;
calculating each characteristic quantity sample based on the remote sensing data and the meteorological observation data of each pixel of each historical remote sensing image sample, and the method comprises the following steps: a first photosynthetically active radiation ratio, a second photosynthetically active radiation ratio, a first moisture limiting factor, a second moisture limiting factor, a first temperature limiting factor, and a second temperature limiting factor;
inputting each characteristic quantity sample into a BP neural network to obtain a vegetation total primary productivity prediction result corresponding to each characteristic quantity sample;
determining a loss function value based on the vegetation total primary productivity prediction result corresponding to each characteristic quantity sample and the corresponding flux tower data;
and updating the weight parameters of the BP neural network based on the loss function values.
In a second aspect, an embodiment of the present application provides an apparatus for estimating total primary productivity of vegetation, including:
the acquisition unit is used for acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period;
the preprocessing unit is used for acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image;
the characteristic quantity calculation unit is used for calculating the characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, and the characteristic quantity comprises the following components: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
and the vegetation total primary productivity estimation unit is used for inputting the characteristic quantity of each pixel of the remote sensing image into a vegetation total primary productivity estimation model which is trained in advance and outputting a vegetation total primary productivity estimation value of each pixel.
In a third aspect, an embodiment of the present application provides an electronic device, including: the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the vegetation total primary productivity estimation method of the embodiment of the application.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the vegetation total primary productivity estimation method of embodiments of the present application.
The method has good adaptability to different vegetation in different regions, can fully reflect the difference of vegetation types and regions, has stronger spatial heterogeneity and better fitting effect, and improves the estimation precision of the total primary productivity of the vegetation.
Drawings
In order to more clearly illustrate the detailed description of the present application or the technical solutions in the prior art, the drawings needed to be used in the detailed description of the present application or the prior art description will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for estimating total primary productivity of vegetation according to an embodiment of the present application;
fig. 2 is a schematic diagram of a training process of a vegetation total primary productivity estimation model provided in an embodiment of the present application;
fig. 3 is a functional block diagram of a vegetation total primary productivity estimation apparatus according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
First, technical terms related to the embodiments of the present application will be briefly described.
Vegetation Total Primary Productivity GPP (gross Primary Productivity): it means the amount of photosynthetic products or the total amount of organic carbon fixed by photosynthesis per unit time of vegetation, also called total primary productivity or total ecosystem productivity.
Remote sensing light energy utilization rate model: the vegetation productivity is estimated by considering only the temperature, the moisture, the illumination and the light energy utilization rate of the area under the condition of simplifying environmental factors.
The Photosynthetically Active radiation absorption ratio FPAR (the Fraction of adsorbed Photosynthetically Active radiation) is generally defined as the absorption ratio of vegetation to solar radiation energy with the wavelength of 400-700 nm, is an important parameter for representing the photosynthesis level and the growth state of the vegetation, and is also one of the key parameters which are identified by the global climate observation system of the United nations and reflect global climate change.
VPM (vacuum chemical photo synthetic model) is a light energy utilization rate model which estimates the total primary productivity by using remote sensing data such as temperature and the like and vorticity observation carbon flux data and considering photosynthetic effective radiation absorbed by vegetation chlorophyll.
CASA (CASA-Ames-Stanford Model) Model is a light energy utilization rate Model which fully considers environmental conditions and vegetation characteristics, and is widely applied to estimation of NPP (Net Primary production) of a land ecosystem.
Cflux (carbon flux) carbon cycle model: a model of light energy utilization using flux stations and remote sensing satellites to estimate total primary productivity.
MOD17 (model Resolution Imaging Spectrophotometer-GPP) model: the method is a light energy utilization rate model for estimating the total primary productivity by mainly utilizing the relationship among land utilization and environmental factors including temperature, water vapor pressure and illumination.
After introducing the technical terms related to the present application, the design ideas of the embodiments of the present application will be briefly described below.
The existing GPP estimation model and algorithm generally have the problem of low precision, and the precision of different global areas in the product is different, so that the actual application requirements are difficult to meet.
In order to solve the technical problem, the historical remote sensing data and the historical meteorological station data are used for calculating to obtain the historical characteristic quantity: photosynthetically active radiation absorption ratio (FPAR), effect of moisture on photosynthesis (moisture limiting factor), and effect of temperature on photosynthesis (temperature limiting factor); training a vegetation total primary productivity estimation model by using the historical characteristic parameters and the historical flux tower data; and processing the current characteristic quantity by using the trained vegetation total primary productivity estimation model to obtain an estimation value of the vegetation total primary productivity. The vegetation total primary productivity estimation model provided by the application depends on the support of machine learning, and compared with a traditional light energy utilization model, the vegetation total primary productivity estimation model has better adaptability to different vegetation in different regions, can better reflect the difference of vegetation types and regions, and has stronger spatial heterogeneity and better fitting effect, so that the vegetation total primary productivity estimation precision is improved. Meanwhile, the process of estimating GPP by using a machine learning algorithm is clearer, and compared with the machine learning of the traditional 'black box' mode, the model has more physical process significance.
Compared with the traditional machine learning of a 'black box' mode, the GPP estimation method introduces a calculation formula with physical significance, so that model training is more targeted, and training efficiency and accuracy can be improved.
After introducing the application scenario and the design concept of the embodiment of the present application, the following describes a technical solution provided by the embodiment of the present application.
As shown in fig. 1, the present application provides a method for estimating total primary productivity of vegetation, comprising:
step 101: acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period;
wherein the monitoring period is determined according to the time resolution of the input data, typically day, month or year.
Step 102: acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image;
because the spatial position corresponding to the meteorological observation data is inconsistent with the spatial position of each pixel of the remote sensing image, interpolation processing needs to be performed on the meteorological observation data to obtain the meteorological observation data at the spatial position of each pixel of the remote sensing image.
Step 103: calculating the characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, wherein the characteristic quantity comprises the following components: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
firstly, calculating the photosynthetically active radiation ratio:
in the VPM model, the photosynthetically active radiation ratio FPAR absorbed by vegetation photosynthesis can be approximately expressed as a linear function of an enhanced vegetation index EVI, the enhanced vegetation index EVI of each pixel of a remote sensing image is obtained, and a first photosynthetically active radiation ratio FPAR is calculatedEVI:
Wherein, in general, the value of the parameter β is 1;
in the CFLUX model and MOD17 model, the photosynthetically active radiation ratio can also be calculated from the Leaf Area Index (LAI).
Obtaining leaf area index of each pixel of remote sensing imageLAICalculating a second photosynthetically active radiation ratio FPARLAI:
Wherein K is the radiation extinction coefficient.
The water limit factor is then calculated:
as the SWIR wave band information is sensitive to the moisture of the earth surface and the vegetation, the NIR wave band is combined to obtain the land surface moisture index LSWI which is sensitive to the moisture content of the leaves, and therefore the first moisture limiting factorW LSWIThe calculation formula of (2) is as follows:
whereinLSWI max Is the maximum LSWI value of vegetation growing season in a single pixel.
In the MOD17 algorithm, the moisture limiting factor may also be derived from the saturated Vapor Pressure Differential (VPD), and the second moisture limiting factor is calculated by the following formulaW VPD:
Wherein the content of the first and second substances,VPD max andVPD min the maximum value and the minimum value of the saturated water vapor pressure difference are related to the vegetation type and can be obtained through an MOD17 model lookup table. When in use𝑉𝑃𝐷 > VPD max When the temperature of the water is higher than the set temperature,W VPDis 0, when𝑉𝑃𝐷 <VPD min When the temperature of the water is higher than the set temperature,W VPDis 1.
Finally, calculating a temperature limiting factor:
using a land ecosystem model (Terrestrial Ecosys)TEM) of the temperature limiting factor, calculating a first temperature limiting factorT TEM :
When in useT min < T < T max The method comprises the following steps:
when in useT ≤T minOrT ≥T max The method comprises the following steps:T TEM = 0
in the formula (I), the compound is shown in the specification,Tis the average temperature (deg.C) over the monitoring period;T opt adopting the temperature corresponding to the maximum EVI of each pixel in the growing season for the temperature corresponding to the maximum photosynthesis rate of the vegetation;T min andT max respectively, the lowest temperature and the highest temperature when the vegetation is subjected to photosynthesis, and when the temperature is lower than the lowest temperature or higher than the highest temperature,T TEM is set to 0.T min The temperature is set to 0 c,T max is calculated by the formulaT opt + ( T opt -T min )2。
The algorithm of the temperature limiting factor in the CASA model is as follows:
T S 1reflecting photosynthesis by biochemical actions in plants at low and high temperaturesAnd the vegetation productivity is reduced, when the temperature is equal to or lower than-10 ℃,T S 1the value is 0, and no photosynthesis occurs at this time;T S 2shows the trend that the light energy utilization rate of the plant gradually decreases when the environment changes from the optimal temperature to the high temperature and the low temperature, for example, the average temperature in a certain monthTSpecific optimum temperatureT opt The average temperature T is 10 deg.C or less than 13 deg.C, and the average temperature T is optimumT opt Substituting the above formula to calculateT S 2A value of which half is taken as the monthly average temperature TT S 2The value of (c).
Step 104: inputting the characteristic quantity of each pixel of the remote sensing image into a pre-trained vegetation total primary productivity estimation model, and outputting a vegetation total primary productivity estimation value of each pixel;
in this embodiment, the vegetation total primary productivity estimation model adopts a BP neural network, and the training process of the vegetation total primary productivity estimation model includes:
step 4A: obtaining a sample set; the system comprises a sample set, a flux tower and a remote sensing image acquisition system, wherein the sample set comprises a plurality of historical remote sensing image samples, corresponding meteorological observation data samples and flux tower data;
60% of the sample data set is used as a training set, 20% is used as a verification set, and 20% is used as a test set.
And step 4B: acquiring spatial position information of each pixel of each historical remote sensing image sample, and rasterizing meteorological observation data to obtain meteorological observation data of each pixel of each historical remote sensing image sample; rasterizing the flux tower data to obtain flux tower data of each pixel of each historical remote sensing image sample;
and step 4C: calculating each characteristic quantity sample based on the remote sensing data and the meteorological observation data of each pixel of each historical remote sensing image sample, and the method comprises the following steps: a first photosynthetically active radiation ratio, a second photosynthetically active radiation ratio, a first moisture limiting factor, a second moisture limiting factor, a first temperature limiting factor, and a second temperature limiting factor;
and step 4D: inputting each characteristic quantity sample into a BP neural network to obtain a vegetation total primary productivity prediction result corresponding to each characteristic quantity sample;
as shown in FIG. 2, first, the initial values of the weights of different layers of the BP neural network are setAnd an initial value of bias(i represents the number of layers);
selecting n samples from a training set; forward propagation: for different layer outputs, according to the following formula:
……
whereinVectors are represented, and weights are represented;bis the bias of the neuron; the activation function f uses a ReLu type function; by usingRepresenting the output of the respective neuron; σ is the GPP prediction result.
And 4E: determining a loss function value based on the vegetation total primary productivity prediction result and flux tower data corresponding to each characteristic quantity sample;
wherein the Loss function L is expressed using Mean-Squared Loss (MSE):
where n is the number of samples,it is shown that the GPP true values, i.e. flux tower data,y pred is a predicted value of GPP. The goal of training the network is to minimize the MSE.
From the above process, when different layer weightsAnd biasbWhen the change occurs, the loss function L is changed, and the change size can be usedAndto indicate.
Due to the fact that
When n = 1, the number of the bits is set to n = 1,
then the process of the first step is carried out,
for theTo demonstrate the calculation process for simplicity, the variability layer is taken to be 0, at this time:
For other weightsAnd biasbThe calculation of the partial derivatives is similar to the above process and will not be described again.
And step 4F: updating the weight parameter of the BP neural network based on the loss function value;
to achieve minimization of the loss function, the weight and bias of the network are optimized using a stochastic gradient descent method. For weight and bias
Where η is a constant used to adjust the speed of the training. When in useWhen the number is positive, the number of the first and second groups is positive,andwill decrease and L will decrease. When each weight and intercept term in the network is so optimized, the loss will continue to decrease and the performance will continue to increase.
And L is continuously reduced along with the continuous updating of the weight and the bias, when the L is at a lower value and the subsequent training L tends to be gentle and does not drop or the drop amplitude is small, the training is stopped, otherwise, the training is continued. In addition, the whole can be manually adjusted, namely, the optimization layer in the flow chart is adjusted, and the optimization layer is added or reducedThe accuracy of the GPP estimation is further improved.
And verifying and testing the trained vegetation total primary productivity estimation model by using a verification set and a test set.
Based on the above embodiments, the present application provides an apparatus for estimating total primary productivity of vegetation, and referring to fig. 3, the apparatus 200 for estimating total primary productivity of vegetation provided by the present application at least includes:
an obtaining unit 201, configured to obtain a remote sensing image and meteorological observation data of a target area in a monitoring time period;
the preprocessing unit 202 is configured to acquire spatial position information of each pixel of the remote sensing image, and perform rasterization processing on meteorological observation data to obtain meteorological observation data of each pixel of the remote sensing image;
the characteristic quantity calculating unit 203 is configured to calculate a characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, where the characteristic quantity includes: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
and a vegetation total primary productivity estimation unit 204, configured to input the feature quantity of each pixel of the remote sensing image into a vegetation total primary productivity estimation model trained in advance, and output a vegetation total primary productivity estimation value of each pixel.
It should be noted that the principle of the vegetation total primary productivity estimation apparatus 200 provided in the embodiment of the present application for solving the technical problem is similar to the vegetation total primary productivity estimation method provided in the embodiment of the present application, and therefore, the implementation of the vegetation total primary productivity estimation apparatus 200 provided in the embodiment of the present application can be referred to the implementation of the vegetation total primary productivity estimation method provided in the embodiment of the present application, and repeated details are not repeated.
As shown in fig. 4, an electronic device 300 provided in the embodiment of the present application at least includes: the vegetation total primary productivity estimation method comprises a processor 301, a memory 302 and a computer program which is stored on the memory 302 and can run on the processor 301, wherein the processor 301 executes the computer program to realize the vegetation total primary productivity estimation method provided by the embodiment of the application.
The electronic device 300 provided by the embodiment of the present application may further include a bus 303 connecting different components (including the processor 301 and the memory 302). Bus 303 represents one or more of any of several types of bus structures, including a memory bus, a peripheral bus, a local bus, and so forth.
The Memory 302 may include readable media in the form of volatile Memory, such as Random Access Memory (RAM) 3021 and/or cache Memory 3022, and may further include Read Only Memory (ROM) 3023.
The memory 302 may also include a program tool 3024 having a set (at least one) of program modules 3025, the program modules 3025 including, but not limited to: an operating subsystem, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
It should be noted that the electronic device 300 shown in fig. 4 is only an example, and should not bring any limitation to the functions and the scope of the application of the embodiments.
The embodiment of the application also provides a computer readable storage medium, and the computer readable storage medium stores computer instructions, and the computer instructions are executed by a processor to realize the vegetation total primary productivity estimation method provided by the embodiment of the application.
It should be noted that although several units or sub-units of the apparatus are mentioned in the above detailed description, such division is merely exemplary and not mandatory. Indeed, the features and functions of two or more units described above may be embodied in one unit, according to embodiments of the application. Conversely, the features and functions of one unit described above may be further divided into embodiments by a plurality of units.
Further, while the operations of the methods of the present application are depicted in the drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.
Claims (8)
1. A method of estimating total primary productivity of vegetation, comprising:
acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period;
acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image;
calculating the characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, wherein the characteristic quantity comprises the following components: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
and inputting the characteristic quantity of each pixel of the remote sensing image into a pre-trained vegetation total primary productivity estimation model, and outputting a vegetation total primary productivity estimation value of each pixel.
2. A method of estimating total primary productivity of vegetation according to claim 1 wherein the step of calculating the proportion of photosynthetically active radiation comprises:
obtaining enhanced vegetation index EVI of each pixel of the remote sensing image, and calculating a first photosynthetically active radiation ratio FPAREVI:
Wherein β is a parameter;
obtaining leaf area index of each pixel of remote sensing imageLAICalculating a second photosynthetically active radiation ratio FPARLAI:
Wherein K is the radiation extinction coefficient.
3. The method of estimating total primary productivity of vegetation according to claim 2 wherein calculating a moisture limiting factor from the remote sensed images and corresponding meteorological observation data comprises:
obtaining land surface moisture index LSWI of each pixel based on SWIR band information and NIR band information of each pixel of the remote sensing image, and calculating a first moisture limiting factorW LSWI:
Wherein the content of the first and second substances,LSWI max is the maximum LSWI value of the vegetation growing season of each pixel element;
acquiring saturated water vapor pressure difference of each pixel of remote sensing image𝑉𝑃𝐷Calculating a second moisture limiting factorW VPD:
Wherein the content of the first and second substances,VPD max andVPD min the maximum value and the minimum value of the saturated water vapor pressure difference are determined according to the vegetation type; when in use𝑉𝑃𝐷 > VPD max When the temperature of the water is higher than the set temperature,W VPDis 0, when𝑉𝑃𝐷 <VPD min When the temperature of the water is higher than the set temperature,W VPDis 1.
4. A method of estimating total primary productivity of vegetation according to claim 3 wherein calculating a temperature limiting factor from the remotely sensed images and corresponding meteorological observation data comprises:
calculating a first temperature limiting factorT TEM :
When in useT min < T < T max The method comprises the following steps:
if not, then,T TEM =0;
in the formula (I), the compound is shown in the specification,Tin order to monitor the average temperature over a period of time,T opt adopting the temperature corresponding to the maximum EVI of each pixel enhanced vegetation index as the temperature corresponding to the maximum photosynthesis rate of vegetation;T min andT max respectively the lowest temperature and the highest temperature when the vegetation is subjected to photosynthesis,T min =0℃,T max =T opt + ( T opt -T min )2;
calculating a second temperature limiting factor using a CASA algorithmT s:
Wherein the content of the first and second substances,T S1in order to reduce the factors of the productivity of vegetation,T S2the trend factor shows that the light energy utilization rate of the plant is gradually reduced.
5. The method of estimating total primary productivity of vegetation of claim 4, wherein the model for estimating total primary productivity of vegetation employs a BP neural network, and wherein the training process of the model for estimating total primary productivity of vegetation comprises:
acquiring a training set; the training set comprises a plurality of historical remote sensing image samples, corresponding meteorological observation data samples and flux tower data;
acquiring spatial position information of each pixel of each historical remote sensing image sample, and rasterizing meteorological observation data to obtain meteorological observation data of each pixel of each historical remote sensing image sample; rasterizing the flux tower data to obtain flux tower data of each pixel of each historical remote sensing image sample;
calculating each characteristic quantity sample based on the remote sensing data and the meteorological observation data of each pixel of each historical remote sensing image sample, and the method comprises the following steps: a first photosynthetically active radiation ratio, a second photosynthetically active radiation ratio, a first moisture limiting factor, a second moisture limiting factor, a first temperature limiting factor, and a second temperature limiting factor;
inputting each characteristic quantity sample into a BP neural network to obtain a vegetation total primary productivity prediction result corresponding to each characteristic quantity sample;
determining a loss function value based on the vegetation total primary productivity prediction result corresponding to each characteristic quantity sample and the corresponding flux tower data;
and updating the weight parameters of the BP neural network based on the loss function values.
6. An apparatus for estimating total primary productivity of vegetation, comprising:
the acquisition unit is used for acquiring remote sensing images and meteorological observation data of a target area in a monitoring time period;
the preprocessing unit is used for acquiring spatial position information of each pixel of the remote sensing image, and rasterizing meteorological observation data to obtain the meteorological observation data of each pixel of the remote sensing image;
the characteristic quantity calculation unit is used for calculating the characteristic quantity of each pixel of the remote sensing image based on the remote sensing data and the meteorological observation data of each pixel of the remote sensing image, and the characteristic quantity comprises the following components: photosynthetically active radiation fraction, moisture limiting factor and temperature limiting factor;
and the vegetation total primary productivity estimation unit is used for inputting the characteristic quantity of each pixel of the remote sensing image into a vegetation total primary productivity estimation model which is trained in advance and outputting a vegetation total primary productivity estimation value of each pixel.
7. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor when executing the computer program implementing the method of estimating total primary productivity of vegetation of any one of claims 1-5.
8. A computer readable storage medium storing computer instructions which, when executed by a processor, implement the method of estimating total primary productivity of vegetation of any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210249453.2A CN114331233B (en) | 2022-03-15 | 2022-03-15 | Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210249453.2A CN114331233B (en) | 2022-03-15 | 2022-03-15 | Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114331233A true CN114331233A (en) | 2022-04-12 |
CN114331233B CN114331233B (en) | 2022-06-24 |
Family
ID=81033923
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210249453.2A Active CN114331233B (en) | 2022-03-15 | 2022-03-15 | Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114331233B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115271991A (en) * | 2022-09-27 | 2022-11-01 | 航天宏图信息技术股份有限公司 | Estimation method and device for net primary productivity of crops |
CN115358095A (en) * | 2022-10-19 | 2022-11-18 | 中国科学院、水利部成都山地灾害与环境研究所 | Method for estimating total primary productivity of high-spatial-resolution vegetation in mountainous region |
CN115601201A (en) * | 2022-10-21 | 2023-01-13 | 中国科学院大气物理研究所(Cn) | Method for calculating total primary productivity of ecological system |
CN116227246A (en) * | 2023-05-09 | 2023-06-06 | 中国科学院、水利部成都山地灾害与环境研究所 | Terrain correction method for vegetation total primary productivity remote sensing product |
CN116432849A (en) * | 2023-04-19 | 2023-07-14 | 哈尔滨工业大学 | Leaf area index prediction method based on primary productivity data assimilation |
CN116628979A (en) * | 2023-05-23 | 2023-08-22 | 中国科学院重庆绿色智能技术研究院 | Multiple regression total primary productivity pixel-by-pixel prediction method, device and medium |
CN116701819A (en) * | 2023-06-13 | 2023-09-05 | 昆明理工大学 | Remote sensing inversion method for total primary productivity of vegetation |
CN118053519A (en) * | 2024-04-16 | 2024-05-17 | 南开大学 | Coupling ResNets and regional vegetation GPP estimation method for improving CASA model |
CN118071020A (en) * | 2024-03-05 | 2024-05-24 | 航天宏图信息技术股份有限公司 | Vegetation climate potential productivity estimation method and device and electronic equipment |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090150126A1 (en) * | 2007-12-10 | 2009-06-11 | Yahoo! Inc. | System and method for sparse gaussian process regression using predictive measures |
CN106599469A (en) * | 2016-12-07 | 2017-04-26 | 浙江农林大学 | Remote sensing monitoring method of light use efficiency and gross primary production of phyllostachys pubescens forest ecological system |
-
2022
- 2022-03-15 CN CN202210249453.2A patent/CN114331233B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090150126A1 (en) * | 2007-12-10 | 2009-06-11 | Yahoo! Inc. | System and method for sparse gaussian process regression using predictive measures |
CN106599469A (en) * | 2016-12-07 | 2017-04-26 | 浙江农林大学 | Remote sensing monitoring method of light use efficiency and gross primary production of phyllostachys pubescens forest ecological system |
Non-Patent Citations (2)
Title |
---|
伍卫星 等: "利用MODIS影像和气候数据模拟中国内蒙古温带草原生态系统总初级生产力", 《中国科学 D辑:地球科学》 * |
我们不生产代码,只是代码的搬运工的博客: "Pandas统计分析基础(数据堆叠+数据清洗)", 《HTTPS://BLOG.CSDN.NET/QQ_50675813/ARTICLE/DETAILS/121439348》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115271991A (en) * | 2022-09-27 | 2022-11-01 | 航天宏图信息技术股份有限公司 | Estimation method and device for net primary productivity of crops |
CN115271991B (en) * | 2022-09-27 | 2023-01-06 | 航天宏图信息技术股份有限公司 | Estimation method and device for net primary productivity of crops |
CN115358095A (en) * | 2022-10-19 | 2022-11-18 | 中国科学院、水利部成都山地灾害与环境研究所 | Method for estimating total primary productivity of high-spatial-resolution vegetation in mountainous region |
US11855453B1 (en) | 2022-10-21 | 2023-12-26 | Institute Of Atmospheric Physics, Chinese Academy Of Sciences | Method for calculating gross primary productivity of ecosystem |
CN115601201A (en) * | 2022-10-21 | 2023-01-13 | 中国科学院大气物理研究所(Cn) | Method for calculating total primary productivity of ecological system |
CN116432849A (en) * | 2023-04-19 | 2023-07-14 | 哈尔滨工业大学 | Leaf area index prediction method based on primary productivity data assimilation |
CN116432849B (en) * | 2023-04-19 | 2024-03-26 | 哈尔滨工业大学 | Leaf area index prediction method based on primary productivity data assimilation |
CN116227246A (en) * | 2023-05-09 | 2023-06-06 | 中国科学院、水利部成都山地灾害与环境研究所 | Terrain correction method for vegetation total primary productivity remote sensing product |
CN116227246B (en) * | 2023-05-09 | 2023-08-11 | 中国科学院、水利部成都山地灾害与环境研究所 | Terrain correction method for vegetation total primary productivity remote sensing product |
CN116628979A (en) * | 2023-05-23 | 2023-08-22 | 中国科学院重庆绿色智能技术研究院 | Multiple regression total primary productivity pixel-by-pixel prediction method, device and medium |
CN116628979B (en) * | 2023-05-23 | 2024-04-30 | 中国科学院重庆绿色智能技术研究院 | Multiple regression total primary productivity pixel-by-pixel prediction method, device and medium |
CN116701819A (en) * | 2023-06-13 | 2023-09-05 | 昆明理工大学 | Remote sensing inversion method for total primary productivity of vegetation |
CN116701819B (en) * | 2023-06-13 | 2024-06-14 | 昆明理工大学 | Remote sensing inversion method for total primary productivity of vegetation |
CN118071020A (en) * | 2024-03-05 | 2024-05-24 | 航天宏图信息技术股份有限公司 | Vegetation climate potential productivity estimation method and device and electronic equipment |
CN118053519A (en) * | 2024-04-16 | 2024-05-17 | 南开大学 | Coupling ResNets and regional vegetation GPP estimation method for improving CASA model |
Also Published As
Publication number | Publication date |
---|---|
CN114331233B (en) | 2022-06-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114331233B (en) | Method and device for estimating total primary productivity of vegetation, electronic equipment and storage medium | |
CN111310889B (en) | Evaporation waveguide profile estimation method based on deep neural network | |
CN108416366B (en) | Power system short-term load prediction method based on meteorological index weighted LS-SVM | |
Liu et al. | Hourly stepwise forecasting for solar irradiance using integrated hybrid models CNN-LSTM-MLP combined with error correction and VMD | |
CN106779154A (en) | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture | |
CN114707688A (en) | Photovoltaic power ultra-short-term prediction method based on satellite cloud chart and space-time neural network | |
CN114510513A (en) | Short-term meteorological forecast data processing method for ultra-short-term photovoltaic power prediction | |
Song et al. | Assessment of solar radiation resource and photovoltaic power potential across China based on optimized interpretable machine learning model and GIS-based approaches | |
CN113297904A (en) | Alpine grassland biomass estimation method and system based on satellite driving model | |
KR20140115777A (en) | Method for estimating longwave climate feedback using sea surface temperature | |
Vimont et al. | Downscaling Indonesian precipitation using large‐scale meteorological fields | |
CN115327662A (en) | Statistical learning-based regional ionized layer TEC prediction method | |
CN113984198A (en) | Short wave radiation prediction method and system based on convolutional neural network | |
TWI717796B (en) | System, method and storage medium for estimating the amount of sunshine in geographic location by artificial intelligence | |
Ramedani et al. | A method based on neural networks for generating solar radiation map | |
CN117293809A (en) | Multi-time space scale new energy generation power prediction method based on large model | |
CN116299247B (en) | InSAR atmospheric correction method based on sparse convolutional neural network | |
CN112560633A (en) | Plant key phenological period time point prediction method and system based on deep learning | |
Li et al. | Machine learning and remote sensing-based modeling of the optimal stomatal behavior of crops | |
CN116776651A (en) | Method and device for measuring and calculating surface evapotranspiration, electronic equipment and storage medium | |
CN116432849A (en) | Leaf area index prediction method based on primary productivity data assimilation | |
CN112036672B (en) | New energy power generation ultra-short term power prediction method and system based on iterative correction | |
CN114047563A (en) | All-weather assimilation method of infrared hyperspectrum | |
Razali et al. | Monitoring green biomass utilizing remote sensing techniques for agriculture and forest areas in East Malaysia | |
CN111752976A (en) | Meteorological data processing method and device, computer equipment and readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |