CN113806943A - Wetland vegetation carbon fixation rate prediction method based on relationship between key water regime variables and vegetation carbon fixation rate - Google Patents
Wetland vegetation carbon fixation rate prediction method based on relationship between key water regime variables and vegetation carbon fixation rate Download PDFInfo
- Publication number
- CN113806943A CN113806943A CN202111106436.5A CN202111106436A CN113806943A CN 113806943 A CN113806943 A CN 113806943A CN 202111106436 A CN202111106436 A CN 202111106436A CN 113806943 A CN113806943 A CN 113806943A
- Authority
- CN
- China
- Prior art keywords
- vegetation
- wetland
- carbon
- carbon fixation
- rate
- 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
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 172
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 172
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000009919 sequestration Effects 0.000 claims abstract description 37
- 230000008859 change Effects 0.000 claims abstract description 22
- 230000004044 response Effects 0.000 claims abstract description 18
- 239000000203 mixture Substances 0.000 claims abstract description 16
- 230000008569 process Effects 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 2
- 238000005259 measurement Methods 0.000 abstract description 7
- 238000011160 research Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000012876 topography Methods 0.000 description 2
- 239000002028 Biomass Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000029553 photosynthesis Effects 0.000 description 1
- 238000010672 photosynthesis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/84—Greenhouse gas [GHG] management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Optimization (AREA)
- Operations Research (AREA)
- Strategic Management (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a wetland vegetation carbon fixation rate prediction method based on a relationship between key water situation variables and vegetation carbon fixation rates, which comprises the following steps: s1, carrying out remote sensing estimation on the vegetation firmness rate of the wetland based on the remote sensing observation vegetation index EVI time sequence image; s2, calculating each hydrological variable which has significance to wetland vegetation by using the wetland flooding duration, the flooding depth, the flooding frequency and the flooding origin-destination time according to the observed water level data and the digital terrain elevation model DEM; s3, calculating a response curve of the vegetation carbon fixation rate to each hydrological variable by using a Gaussian mixture model, and obtaining a key water situation variable with the highest interpretation degree on the wetland vegetation carbon fixation rate; and S4, predicting the carbon sequestration rate of the wetland vegetation after the change of the key water regimen variables through the comparison of the key water regimen variables in the prediction stage and the observation stage. The method is suitable for the wetland ecosystem with higher water level variation, which has less ground carbon stock actual measurement data and plays a leading role in the water regime on the vegetation, and has simple and convenient operation method, and convenient popularization and application.
Description
Technical Field
The invention relates to the technical field of wetland auxiliary estimation, in particular to a wetland vegetation carbon fixation rate prediction method based on the relationship between key water regime variables and vegetation carbon fixation rates.
Background
Carbon elements fixed by wetland plants through photosynthesis, namely plant biomass carbon, are the main carbon input in the carbon fixation process of the wetland ecosystem. The carbon storage quantity accumulated and increased amount of the wetland vegetation in unit time, namely the carbon sequestration rate of the wetland vegetation directly reflects the carbon sink state of the wetland ecosystem and is a main factor concerned by the adjustment of the wetland ecological process. The hydrological conditions are main control factors of the wetland ecosystem for influencing the vegetation, and various hydrological variables including flooding depth, flooding duration and the like all have important influence on the carbon sequestration rate of the wetland vegetation. In recent years, the water and texture situations of a plurality of shallow lakes are remarkably changed due to the accelerated climate change and human activities, and the carbon sequestration function of wetland vegetation is further threatened. In this context, the assessment of the change in the carbon sequestration rate of wetland vegetation in changing environments has gradually become the focus of ecological hydrologic attention.
Although research has been carried out to develop methods for estimating the carbon fixation rate of vegetation in a wide range of wetlands by using remote sensing means, estimation of the carbon fixation rate of the wetland vegetation based on the remote sensing means often requires calibration and verification of measured vegetation carbon stock data on the ground. And the carbon storage ground observation sequence is often short, so that the carbon fixation rate of the wetland vegetation is difficult to estimate by adopting a remote sensing method in the period without ground carbon storage observation data, and the long-acting influence mechanism evaluation of factors such as climate change or water conservancy engineering on the carbon fixation rate of the wetland vegetation cannot be realized.
Generally, the observation sequence of the hydrological data is more complete and longer, for example, the hydrological observation data of China mostly starts from the beginning of building the country, and the hydrological data sequence of 1950s has already become complete. And a plurality of researches show that the carbon sequestration rate of the wetland vegetation depends on the hydrological conditions to a great extent, so that the carbon sequestration rate of the wetland vegetation in the carbon-stock-free ground observation data period can be estimated based on the correlation between the key water regime variable and the carbon sequestration rate of the wetland vegetation, and the result is favorable for the long-acting influence mechanism research of factors such as climate change or water conservancy engineering on the carbon sequestration rate of the wetland vegetation.
In view of this, it is necessary to provide a method for estimating carbon fixation rate of wetland vegetation based on the relationship between key water regime variables and carbon fixation rate of vegetation for the purpose of evaluating long-term influence mechanisms on carbon fixation rate of wetland vegetation, such as climate change or water conservancy engineering.
Disclosure of Invention
The invention aims to solve the technical problem of providing a wetland vegetation carbon fixation rate prediction method based on the relationship between key water regime variables and vegetation carbon fixation rates, which is suitable for a wetland ecosystem with small actual measurement data of ground carbon stock and high water level amplitude and water regime dominating vegetation, and has the advantages of simple and convenient operation method, and convenience for popularization and application.
In order to solve the technical problem, the invention provides a wetland vegetation carbon fixation rate prediction method based on the relationship between key water regime variables and vegetation carbon fixation rates, which comprises the following steps:
s1, carrying out remote sensing estimation on the vegetation firmness rate of the wetland based on the remote sensing observation vegetation index EVI time sequence image;
s2, calculating each hydrological variable which has significance to wetland vegetation by using the wetland flooding duration, the flooding depth, the flooding frequency and the flooding origin-destination time according to the observed water level data and the digital terrain elevation model DEM;
s3, calculating a response curve of the vegetation carbon fixation rate to each hydrological variable by using a Gaussian mixture model, and obtaining a key water situation variable with the highest interpretation degree on the wetland vegetation carbon fixation rate;
and S4, predicting the carbon sequestration rate of the wetland vegetation after the change of the key water regimen variables through the comparison of the key water regimen variables in the prediction stage and the observation stage.
Preferably, in step S1, the original EVI time-series images are processed by S-G filtering to obtain a reconstructed high-quality EVI image time-series with high spatial and temporal consistency; the method comprises the following steps of taking wetland vegetation carbon stock sample observation data as ground verification, realizing remote sensing inversion of an annual dynamic change process of the wetland vegetation carbon stock, and calculating annual carbon stock accumulated growth amount of the wetland vegetation based on an accumulated stock method, namely the carbon sequestration rate of the wetland vegetation;
the basic operation process of interpolating the abnormal value of the original EVI image by S-G filtering is as follows: selecting the values of m adjacent points near the EVI image abnormal point j to fit a d-order polynomial to obtain the polynomial coefficient C of each pointiTo predict the fitted value at point j. The size of the sliding window width m is determined by the length of the EVI time sequence image, polynomial orderNumber d and S-G filter coefficient CiAll determined by the least square method;
and performing remote sensing inversion on the carbon storage of the wetland vegetation by the single reconstructed EVI image to determine an optimal wetland vegetation carbon storage remote sensing inversion model in a linear regression equation, a binomial equation, an index model or a random forest algorithm of the measured carbon storage and the EVI value of the corresponding coordinate position at the same period. The annual carbon sequestration rate of the wetland vegetation is calculated by an accumulation stock method, namely, the accumulated growth amount of the carbon stocks of the wetland vegetation among the carbon stocks of the wetland vegetation in each year is calculated, and the annual carbon sequestration rate of the wetland vegetation is obtained.
Preferably, in step S2, the average daily water level of the multi-hydrological station on the lake surface is used as the average water level of the lake area, the average daily water level is rasterized to obtain time series data of the average daily water level of the lake, and the difference between the time series data and the digital elevation model DEM of the lake is calculated to obtain a spatial distribution map of the daily flooding state of the lake; on the basis, the annual Inundation Duration (IDU), the Average Inundation Depth (AID), the First Inundation Start time (SFI) and the Last Inundation End time (ELI) of each grid are counted to obtain a multi-water-condition variable space distribution map with significant ecological significance to wetland vegetation;
the flooding depth is an average value of daily flooding depths of annual flooding time periods, and is calculated according to the following formula:
in the formula, WSEtAnd the elevation of the lake water surface on the t day is illustrated by elev, and the topography of the lake basin is illustrated by the lake basin elevation in the digital elevation model DEM.
Preferably, in step S3, describing and quantifying a distribution pattern of the wetland vegetation carbon fixation rate along each water situation variable gradient by using a Gaussian Mixture Model (GMM), and revealing a response behavior of the wetland vegetation carbon fixation rate to the lake water situation; comparing the interpretation degree of each regimen variable on the carbon fixation rate of the wetland vegetation, selecting the variable with the highest interpretation degree on the carbon fixation rate of the wetland vegetation as a key regimen variable to obtain each parameter value estimated on the carbon fixation rate of the wetland vegetation, and establishing a GMM (Gaussian mixture model) -based key regimen variable-wetland vegetation solid rate prediction model;
establishing a relation model of a plurality of water situation variables to the carbon storage of wetland vegetation by using a Gaussian mixture model GMM to determine a coefficient/goodness of fit R2And selecting R as the fitting effect of the selection standard to each model by taking the root mean square error/effective value RMSE as the selection standard2The water regime variable in the model with the highest RMSE and the lowest RMSE is taken as a key water regime variable, and a relation model between the key water regime variable and the carbon fixation rate of the wetland vegetation is taken as an optimal wetland vegetation carbon fixation rate prediction model; wherein, the general form of the Gaussian mixture model is as follows:
wherein the variable y is the carbon sequestration rate of the wetland vegetation; the variable x is a water regime condition, and comprises four water regime variables of flooding duration, initial flooding starting time, final flooding ending time and average flooding depth; parameter c1、c2Two peak values of the carbon sequestration rate of the wetland vegetation are respectively; u. of1、u2Respectively taking values of the regimen variable when the carbon sequestration rate of the wetland vegetation reaches a first peak value and a second peak value; t is t1、t2The peak widths of the two peaks are respectively used for describing the fluctuation range of the water regime variables for ensuring the normal carbon fixation rate of the wetland vegetation.
Preferably, in step S4, dividing key regimen variables into equal difference sequences with appropriate step length, establishing homogeneous hydrological response units with wetland vegetation carbon fixation rate based on the key regimen variables, counting frequency distribution maps of the key regimen variables in observation periods and prediction periods with the homogeneous hydrological response units, comparing differences of the key regimen variables in two periods, calculating wetland vegetation carbon fixation rate in the prediction periods respectively in each homogeneous hydrological response unit, and revealing wetland vegetation carbon fixation rate change caused by the key regimen change in two periods through comparison of prediction results and vegetation carbon fixation rate in actual measurement periods; the key water regime variable step length for dividing the homogeneous hydrological response units is generally 1/n of the value range of the key water regime variable, the higher the value of n is, the higher the precision of the prediction result is, and the value of n is between 15 and 20 according to the balance between the prediction precision and the calculated amount.
The invention has the beneficial effects that: the method can effectively predict the carbon fixation rate of the vegetation in the wetland in the period of no ground carbon stock actual measurement data, discriminate the change process of the carbon fixation rate of the vegetation in the wetland caused by the change of key regimen variables, and achieve the purposes of estimating the carbon stock of the vegetation in the wetland in the period of no ground observation and quantifying the influence of the key regimen changes on the carbon fixation rate of the vegetation in the wetland by combining three methods of key regimen variable recognition, establishing a relation curve between the key regimen variables and the carbon fixation rate of the vegetation in the wetland and comparing the observation period with the key regimen variables in the prediction period; the operation is simple and convenient, and the required cost is low.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
Fig. 2 is a schematic diagram of estimating the carbon sequestration rate of the Poyang lake wetland vegetation based on the remote sensing observation vegetation index time-series image in embodiment 1 of the invention.
FIG. 3 is a spatial distribution diagram of the flow situation variables of the Poyang lake wetland calculated in embodiment 1 of the present invention.
Fig. 4 is a schematic diagram of a model of the relationship between the water regime variables and the carbon fixation rate of wetland vegetation obtained in embodiment 1 of the present invention.
Fig. 5(a) is a comparison graph of the distribution areas of the key water regime variables of the observation period and the prediction period in example 1 of the present invention.
Fig. 5(b) is a graph of the difference in carbon sequestration rate of wetland vegetation estimated from the difference in water conditions between the observation period and the prediction period in example 1 of the present invention.
Detailed Description
As shown in fig. 1, a method for predicting carbon fixation rate of wetland vegetation based on the relationship between key water situation variables and carbon fixation rate of vegetation comprises the following steps:
s1, carrying out remote sensing estimation on the vegetation natural rate of the wetland based on the remote sensing observation vegetation index (EVI) time sequence image;
s2, calculating each hydrological variable which has significance to wetland vegetation, such as wetland flooding duration, flooding depth, flooding frequency, flooding origin-destination time and the like by using the observed water level data and a digital terrain elevation model (DEM);
s3, calculating a response curve of the vegetation carbon fixation rate to each hydrological variable by using a Gaussian mixture model, and obtaining a key water situation variable with the highest interpretation degree on the wetland vegetation carbon fixation rate;
and S4, predicting the carbon sequestration rate of the wetland vegetation after the change of the key water regimen variables through the comparison of the key water regimen variables in the prediction stage and the observation stage.
The method specifically comprises the following steps:
firstly, remote sensing observation EVI time sequence images and ground measured carbon storage data are used for estimating the carbon sequestration rate of wetland vegetation in a research area; meanwhile, water situation variables such as the duration of flooding, the depth of flooding and the beginning-end time of flooding in the research area are calculated according to the water level observation data and the DEM data; on the basis, a relation model of each water situation variable and the carbon fixation rate of the wetland vegetation is established by using a Gaussian mixture model GMM, a prediction variable of an optimal model is identified as a key water situation variable, the difference between an observation time period and a prediction time period is compared, and the carbon fixation rate of the wetland vegetation in a carbon-free stock sample actual measurement data time period is predicted by using the difference.
Specifically, the method comprises the following specific technical scheme:
s1, processing the original EVI time sequence image by S-G filtering to obtain a reconstructed high-quality EVI image time sequence with high space-time consistency; and taking the wetland vegetation carbon stock sample observation data as ground verification, realizing remote sensing inversion of the annual dynamic change process of the wetland vegetation carbon stock based on the reconstructed high-quality EVI image, and calculating the annual carbon stock accumulated growth amount of the wetland vegetation based on an accumulated stock method, namely the carbon fixation rate of the wetland vegetation.
The basic operation process of interpolating the abnormal value of the original EVI image by S-G filtering is as follows: selecting the values of m adjacent points near the EVI image abnormal point j to fit a d-order polynomial to obtain the polynomial coefficient C of each pointiTo predict the fitted value at point j. The sliding window width m is determined by the length of the EVI time sequence image, the polynomial order d and the S-G filter coefficient CiAre determined by the least squares method.
And performing remote sensing inversion on the carbon storage of the wetland vegetation by the single reconstructed EVI image to determine an optimal wetland vegetation carbon storage remote sensing inversion model in a linear regression equation, a binomial equation, an index model or a random forest algorithm of the measured carbon storage and the EVI value of the corresponding coordinate position at the same period. The annual carbon sequestration rate of the wetland vegetation is calculated by an accumulation stock method, namely, the accumulated growth amount of the carbon stocks of the wetland vegetation among the carbon stocks of the wetland vegetation in each year is calculated, and the annual carbon sequestration rate of the wetland vegetation is obtained.
S2, taking the average daily water level of the multi-hydrological station on the lake surface as the average water level of the lake area, rasterizing the average daily water level to obtain time sequence data of the average daily water level of the lake, and calculating the difference between the time sequence data and the digital elevation model DEM of the lake to obtain the daily flooding state space distribution map of the lake. On the basis, the number of days that each grid elevation is lower than the lake water level is counted as the flooding duration IDU, the average value of the flooding depths in all flooding time periods in each grid year is used as the flooding depth AID, the initial flooding start time SFI and the last flooding end time ELI, and the multi-water-plot-variable space distribution map which has significant ecological significance on wetland vegetation is obtained.
The flooding depth is an average value of daily flooding depths of annual flooding time periods, and is calculated according to the following formula:
in the formula, WSEtAnd the elevation of the lake water surface on the t day is illustrated by elev, and the topography of the lake basin is illustrated by the lake basin elevation in the digital elevation model DEM.
S3, establishing a relation model of a plurality of water situation variables to wetland vegetation carbon storage by using the Gaussian mixture model GMM to determine coefficient/goodness of fit R2And selecting R as the fitting effect of the selection standard to each model by taking the root mean square error/effective value RMSE as the selection standard2And taking the water regime variable in the model with the highest RMSE as a key water regime variable, and taking a relation model of the water regime variable and the carbon fixation rate of the wetland vegetation as an optimal wetland vegetation carbon fixation rate prediction model. Wherein, the general form of the Gaussian mixture model is as follows:
wherein the variable y is the carbon sequestration rate of the wetland vegetation; the variable x is a water situation condition, and comprises four water situation variables of flooding duration, initial flooding starting time, final flooding ending time and average flooding depth. Parameter c1、c2Two peak values of the carbon sequestration rate of the wetland vegetation are respectively; u. of1、u2Respectively taking values of the regimen variable when the carbon sequestration rate of the wetland vegetation reaches a first peak value and a second peak value; t is t1、t2The peak widths of the two peaks are respectively used for describing the fluctuation range of the water regime variables for ensuring the normal carbon fixation rate of the wetland vegetation.
S4, dividing key water condition variables into equal difference sequences according to proper step length, establishing homogeneous hydrological response units of the wetland vegetation carbon fixation rate based on the key water condition variables, counting frequency distribution graphs of the key water condition variables in the observation period and the prediction period according to the homogeneous hydrological response units, comparing the difference of the key water condition variables in the two periods, respectively calculating the wetland vegetation carbon fixation rate in the prediction period in each homogeneous hydrological response unit, and revealing the change of the wetland vegetation carbon fixation rate caused by the key water condition change in the two periods through the comparison of the prediction result and the vegetation carbon fixation rate in the actual measurement period. The key water regime variable step length for dividing the homogeneous hydrological response units is generally 1/n of the value range of the key water regime variable, the higher the value of n is, the higher the precision of the prediction result is, and n is generally between 15 and 20 according to the balance between the prediction precision and the calculated amount.
The present invention is further illustrated by the following specific examples.
Example 1:
taking a typical wetland ecosystem Poyang lake wetland as an example.
FIG. 2 is a schematic diagram of estimating the carbon fixation rate of the Poyang lake wetland vegetation based on remote sensing observation of vegetation index time sequence images. The method specifically comprises the steps that firstly, according to an AQ value (AQ) of a general Quality image layer (Band Pixel Reliability), low-Quality pixels with AQ values of-1, 2 and 3 are deleted, and only middle-Quality and high-Quality pixels are reserved; fitting the EVI time sequence curve by S-G filtering to complete the interpolation of the removed abnormal value (figure 2 a); secondly, establishing a functional relation between the measured wetland vegetation carbon storage and the EVI based on a binomial expression, establishing a remote sensing inversion model (figure 2b) of the wetland vegetation carbon storage, inverting the wetland vegetation carbon storage corresponding to the remote sensing images in each period by using the remote sensing inversion model, and calculating the annual accumulated growth amount of the carbon storage by using an accumulated storage method to obtain the carbon sequestration rate of the wetland vegetation, as shown in figure 2 c.
FIG. 3 shows the average lake flooding depth AID, flooding duration IDU, flooding start time SFI and flooding end time ELI of Poyang lake 2003-2016 calculated from the lake day level observation data and DEM data over a plurality of years. As can be seen from fig. 3, the spatial distribution of each regimen variable of the wetland can be well calculated by combining the water level observation value and the spatial statistical method of the lake basin terrain data, and each regimen variable shows a good gradient change from the lake center to the lake bank.
FIG. 4 is a relation curve of each water situation variable and the carbon sequestration rate of wetland vegetation based on Gaussian mixture model GMM fitting. As can be seen from fig. 4: the interpretation degree of the waterflooding depth AID on the space change of the carbon sequestration rate of the wetland vegetation is the highest, and the waterflooding depth AID reaches 92 percent and is higher than the waterflooding duration IDU (91 percent), the waterflooding starting time SFI (89 percent) and the waterflooding ending time ELI (91 percent). Therefore, the waterflood depth AID is a key water situation variable for determining the carbon fixing rate of the wetland vegetation in the Poyang lake, and the estimation of the carbon fixing rate of the wetland vegetation can be realized by the waterflood depth AID.
FIGS. 5(a) and 5(b) are the comparison between the key water regime variable AID in the observation period (2003-. It can be seen that the carbon fixation rate of the vegetation increases with the increase of the area with low flooding depth, and similarly, the carbon fixation rate of the vegetation decreases with the decrease of the area with high flooding depth, and overall, the carbon fixation rate of the vegetation increases more than the decrease, so that the carbon fixation rate of the wetland vegetation increases in 2003-2016 in total in comparison with 1980-2002.
The method realizes the purpose of predicting the carbon fixation rate of the wetland vegetation in the period of actual carbon stock data loss by utilizing an MODIS (modified Resolution Imaging spectrometer, MODIS) image time sequence, observation water level data, a research area digital elevation model and ground carbon stock actual measurement data, identifying key water situation variables, establishing a functional relation between the key water situation variables and the carbon fixation rate of the wetland vegetation, and comparing the key water situation variables in the observation period and the prediction period. The method quantifies the influence of key water regime variables on the carbon sequestration rate of the wetland vegetation, predicts the carbon sequestration rate of the wetland vegetation in the observation period without carbon storage, and has obvious advantages and innovativeness.
Claims (5)
1. A wetland vegetation carbon fixation rate prediction method based on a relationship between key water situation variables and vegetation carbon fixation rates is characterized by comprising the following steps:
s1, carrying out remote sensing estimation on the vegetation firmness rate of the wetland based on the remote sensing observation vegetation index EVI time sequence image;
s2, calculating each hydrological variable which has significance to wetland vegetation by using the wetland flooding duration, the flooding depth, the flooding frequency and the flooding origin-destination time according to the observed water level data and the digital terrain elevation model DEM;
s3, calculating a response curve of the vegetation carbon fixation rate to each hydrological variable by using a Gaussian mixture model, and obtaining a key water situation variable with the highest interpretation degree on the wetland vegetation carbon fixation rate;
and S4, predicting the carbon sequestration rate of the wetland vegetation after the change of the key water regimen variables through the comparison of the key water regimen variables in the prediction stage and the observation stage.
2. The wetland vegetation carbon fixation rate prediction method based on the relationship between the key water situation variables and the vegetation carbon fixation rate of claim 1, wherein in step S1, the original EVI time series images are processed by S-G filtering to obtain a reconstructed high-quality EVI image time series with high space-time consistency; the method comprises the following steps of taking wetland vegetation carbon stock sample observation data as ground verification, realizing remote sensing inversion of an annual dynamic change process of the wetland vegetation carbon stock, and calculating annual carbon stock accumulated growth amount of the wetland vegetation based on an accumulated stock method, namely the carbon sequestration rate of the wetland vegetation;
the basic operation process of interpolating the abnormal value of the original EVI image by S-G filtering is as follows: selecting the values of m adjacent points near the EVI image abnormal point j to fit a d-order polynomial to obtain the polynomial coefficient C of each pointiThe fitting value at the prediction point j, the size of the sliding window width m is determined by the length of the EVI time sequence image, the polynomial order d and the S-G filter coefficient CiAll determined by the least square method;
remote sensing inversion of the single reconstructed EVI image on the carbon storage of the wetland vegetation is carried out to determine an optimal remote sensing inversion model of the carbon storage of the wetland vegetation in a linear regression equation, a binomial equation, an index model or a random forest algorithm of the actually measured carbon storage and the EVI value of the corresponding coordinate position at the same period; the annual carbon sequestration rate of the wetland vegetation is calculated by an accumulation stock method, namely, the accumulated growth amount of the carbon stocks of the wetland vegetation among the carbon stocks of the wetland vegetation in each year is calculated, and the annual carbon sequestration rate of the wetland vegetation is obtained.
3. The method for predicting the carbon fixation rate of the wetland vegetation based on the relationship between the key water situation variables and the carbon fixation rate of the vegetation, as claimed in claim 1, wherein in step S2, the average daily water level of the multiple hydrological stations on the lake surface is used as the average water level of the lake area, the average daily water level of the lake area is rasterized to obtain time series data of the average daily water level of the lake, and the difference between the time series data and the digital elevation model DEM of the lake is calculated to obtain the spatial distribution map of the daily submerging state of the lake; on the basis, the annual flooding duration IDU, the average flooding depth AID, the initial flooding start time SFI and the last flooding end time ELI of each grid are counted to obtain a multi-water-condition-variable spatial distribution map which has significant ecological significance on wetland vegetation;
the flooding depth AID is an average value of flooding depths of days in an annual flooding period, and is calculated according to the following formula:
in the formula, WSEtIs the lake surface elevation of the t day, elev is the lake basin terrain, namely the lake basin elevation in the digital elevation model DEM。
4. The method for predicting the carbon fixation rate of the wetland vegetation based on the relationship between the key water situation variables and the carbon fixation rate of the vegetation in the claim 1, wherein in the step S3, a Gaussian mixture model GMM is used for describing and quantifying the distribution mode of the carbon fixation rate of the wetland vegetation along the gradient of each water situation variable, so as to reveal the response behavior of the carbon fixation rate of the wetland vegetation to the hydrological situation of the lake; comparing the interpretation degree of each regimen variable on the carbon fixation rate of the wetland vegetation, selecting the variable with the highest interpretation degree on the carbon fixation rate of the wetland vegetation as a key regimen variable to obtain each parameter value estimated on the carbon fixation rate of the wetland vegetation, and establishing a GMM (Gaussian mixture model) -based key regimen variable-wetland vegetation solid rate prediction model;
establishing a relation model of a plurality of water situation variables to the carbon storage of wetland vegetation by using a Gaussian mixture model GMM to determine a coefficient/goodness of fit R2And selecting R as the fitting effect of the selection standard to each model by taking the root mean square error/effective value RMSE as the selection standard2The water regime variable in the model with the highest RMSE and the lowest RMSE is taken as a key water regime variable, and a relation model between the key water regime variable and the carbon fixation rate of the wetland vegetation is taken as an optimal wetland vegetation carbon fixation rate prediction model; wherein, the general form of the Gaussian mixture model is as follows:
wherein the variable y is the carbon sequestration rate of the wetland vegetation; the variable x is a water regime condition, and comprises four water regime variables of flooding duration, initial flooding starting time, final flooding ending time and average flooding depth; parameter c1、c2Two peak values of the carbon sequestration rate of the wetland vegetation are respectively; u. of1、u2Respectively taking values of the regimen variable when the carbon sequestration rate of the wetland vegetation reaches a first peak value and a second peak value; t is t1、t2The peak widths of the two peaks are respectively used for describing the fluctuation range of the water regime variables for ensuring the normal carbon fixation rate of the wetland vegetation.
5. The method for predicting the carbon fixation rate of the wetland vegetation based on the relationship between the key water condition variables and the carbon fixation rate of the vegetation according to claim 1, wherein in step S4, the key water condition variables are divided into equal difference sequences by proper step length, homogeneous hydrological response units of the wetland vegetation carbon fixation rate based on the key water condition variables are established, frequency distribution maps of the key water condition variables in the observation period and the prediction period are counted by the homogeneous hydrological response units, the difference of the key water condition variables in the two periods is compared, the carbon fixation rate of the wetland vegetation in the prediction period is respectively calculated in each homogeneous hydrological response unit, and the change of the carbon fixation rate of the wetland vegetation caused by the key water condition change in the two periods is revealed by the comparison of the prediction result and the actually measured carbon fixation rate of the vegetation in the two periods; the key water regime variable step length for dividing the homogeneous hydrological response units is generally 1/n of the value range of the key water regime variable, the higher the value of n is, the higher the precision of the prediction result is, and the value of n is between 15 and 20 according to the balance between the prediction precision and the calculated amount.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111106436.5A CN113806943B (en) | 2021-09-22 | 2021-09-22 | Wetland vegetation carbon fixation rate prediction method based on relation between key water regime variables and vegetation carbon fixation rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111106436.5A CN113806943B (en) | 2021-09-22 | 2021-09-22 | Wetland vegetation carbon fixation rate prediction method based on relation between key water regime variables and vegetation carbon fixation rate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113806943A true CN113806943A (en) | 2021-12-17 |
| CN113806943B CN113806943B (en) | 2024-04-02 |
Family
ID=78939851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111106436.5A Active CN113806943B (en) | 2021-09-22 | 2021-09-22 | Wetland vegetation carbon fixation rate prediction method based on relation between key water regime variables and vegetation carbon fixation rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113806943B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082274A (en) * | 2022-06-09 | 2022-09-20 | 贵州师范学院 | Earth surface vegetation carbon sink estimation and transaction method and system based on satellite remote sensing |
| CN117969789A (en) * | 2024-03-29 | 2024-05-03 | 山东省地矿工程勘察院(山东省地质矿产勘查开发局八〇一水文地质工程地质大队) | Water regime prediction method and monitoring system for underground water resource |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359411A (en) * | 2018-11-01 | 2019-02-19 | 中国科学院东北地理与农业生态研究所 | A kind of Marsh Wetland vegetation fraction estimation method under climate change effect |
| CN112991226A (en) * | 2021-04-15 | 2021-06-18 | 河海大学 | Remote sensing estimation method for carbon sequestration rate of wetland vegetation based on reconstructed vegetation index time sequence image |
-
2021
- 2021-09-22 CN CN202111106436.5A patent/CN113806943B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109359411A (en) * | 2018-11-01 | 2019-02-19 | 中国科学院东北地理与农业生态研究所 | A kind of Marsh Wetland vegetation fraction estimation method under climate change effect |
| CN112991226A (en) * | 2021-04-15 | 2021-06-18 | 河海大学 | Remote sensing estimation method for carbon sequestration rate of wetland vegetation based on reconstructed vegetation index time sequence image |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082274A (en) * | 2022-06-09 | 2022-09-20 | 贵州师范学院 | Earth surface vegetation carbon sink estimation and transaction method and system based on satellite remote sensing |
| CN117969789A (en) * | 2024-03-29 | 2024-05-03 | 山东省地矿工程勘察院(山东省地质矿产勘查开发局八〇一水文地质工程地质大队) | Water regime prediction method and monitoring system for underground water resource |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113806943B (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113806943A (en) | Wetland vegetation carbon fixation rate prediction method based on relationship between key water regime variables and vegetation carbon fixation rate | |
| Adarsh et al. | Trend analysis of rainfall in four meteorological subdivisions of southern India using nonparametric methods and discrete wavelet transforms. | |
| Coopersmith et al. | Soil moisture model calibration and validation: An ARS watershed on the South Fork Iowa River | |
| Bhuyan et al. | Regional flood frequency analysis of north-bank of the river Brahmaputra by using LH-moments | |
| Omane-Adjepong et al. | Determining the better approach for short-term forecasting of ghana’s inflation: Seasonal ARIMA Vs holt-winters | |
| CN111242404A (en) | Extreme evaluation method and system for heavy rainfall induced flood incident | |
| Simpson et al. | Analysis of UK precipitation extremes derived from Met Office gridded data | |
| CN114297578A (en) | Grassland vegetation coverage estimation and prediction method based on remote sensing | |
| Radziejewski et al. | Fractal analysis of flow of the river Warta | |
| CN110008439B (en) | Rainfall data space-time integrated interpolation algorithm based on matrix decomposition | |
| Mahbod et al. | A comparative analysis of the precipitation extremes obtained from tropical rainfall-measuring mission satellite and rain gauges datasets over a semiarid region | |
| CN107729651A (en) | Domatic rill developmental morphology characteristic synthetic quantization method based on various dimensions | |
| CN104143031B (en) | A kind of vegetation index time series data reconstructing method based on Multiscale Wavelet Decomposition | |
| CN107831516B (en) | Method for acquiring real-time high-precision displacement of dam by fusing GNSS and ground monitoring network | |
| CN112991226B (en) | Remote sensing estimation method for carbon sequestration rate of wetland vegetation based on reconstructed vegetation index time sequence image | |
| Rashid et al. | Statistical characteristics of rainfall in the Onkaparinga catchment in South Australia | |
| Acquaotta et al. | Estimation of rainfall erosivity in Piedmont (Northwestern Italy) by using 10-minute fixed-interval rainfall data | |
| CN115329610B (en) | Method, device and equipment for identifying drought and waterlogging emergency turn based on soil moisture | |
| Sivakumar et al. | Nonlinear analysis of rainfall dynamics in California's Sacramento Valley | |
| CN115660134A (en) | Flood storage regulation spatialization method and system based on remote sensing data | |
| CN115310796A (en) | Method and system for determining propagation relationship among different types of drought | |
| Barbosa et al. | Nonlinear sea level trends from European tide gauge records | |
| CN111428932B (en) | Medium-and-long-term air traffic flow prediction method based on wavelet transformation and gray prediction | |
| CN109522520B (en) | Multiple wavelet coherence analysis method for underground water level fluctuation and multiple factors | |
| CN111723479A (en) | Process-oriented eutrophic lake algae total quantity remote sensing estimation method |
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 |