A method for accounting for carbon sinks of rivers and lakes based on remote sensing data

Abstract

The application discloses a kind of river and lake carbon sink accounting methods based on remote sensing data, belong to environmental monitoring and remote sensing technical field.To solve the problem that traditional means is difficult to evaluate static carbon storage and dynamic flux synergistically, the application obtains multi-source remote sensing image and synchronous meteorological data;Based on the water column total carbon storage obtained by the inversion of water body organic carbon concentration and the combination of water depth;Inversion water surface temperature and chlorophyll-a concentration to determine water carbon dioxide partial pressure, calculate water-gas interface carbon dioxide exchange flux;Finally, a synergistic accounting equation is established, coupling cumulative exchange flux, carbon storage change and exogenous carbon input, deducting exogenous input to output endogenous net carbon sink amount.The application realizes large-scale, high-precision dynamic monitoring of river and lake carbon sink, significantly improves the accuracy of water source-sink attribute identification.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring and remote sensing information data processing technology, specifically relating to a method for calculating carbon sequestration in rivers and lakes based on remote sensing data. Background Technology

[0002] Lakes and rivers, as important components of the terrestrial hydrosphere, play a crucial role in the global carbon cycle. Accurately calculating the carbon storage and carbon flux of rivers and lakes is of great significance for assessing the carbon source / sink properties of regions and formulating carbon neutrality policies.

[0003] However, existing methods for monitoring and accounting for carbon sinks in rivers and lakes have significant shortcomings. On the one hand, traditional methods rely heavily on in-situ contact sampling, which is limited by manpower, resources, and extreme weather, making it impossible to achieve large-scale, periodic, and continuous spatial dynamic monitoring. On the other hand, existing technologies often isolate the carbon sink process in water bodies, only assessing the static carbon storage or the instantaneous carbon dioxide exchange flux at the water-air interface in isolation. This makes it difficult to systematically couple the spatiotemporal integral of dynamic flux with changes in carbon storage, and lacks effective multi-temporal data closed-loop calculation logic. More seriously, existing accounting systems typically ignore the severe interference from external carbon inputs to the watershed and lack inversion methods that couple remote sensing and hydrological data to obtain accurate external fundamental parameters. This results in calculations that fail to accurately reflect the endogenous net carbon sink generated by biogeochemical processes within the water body, leading to significant biases in the final determination of net carbon source and sink attributes.

[0004] Currently, there is a lack of a suitable technology that can effectively solve the above-mentioned problems. The information disclosed in this background section is intended only to enhance the understanding of the general background of the invention and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] To address the shortcomings of existing technologies, such as the difficulty in coordinating the assessment of static carbon storage and dynamic flux changes in rivers and lakes, and the lack of effective multi-temporal closed-loop calculation logic to accurately isolate exogenous carbon inputs, this invention provides a river and lake carbon sink accounting method based on remote sensing data to achieve a rigorous assessment of endogenous net carbon sinks and accurately determine the net carbon source and sink attributes of water bodies.

[0006] The technical solution of the present invention is as follows: A method for calculating carbon sinks in rivers and lakes based on remote sensing data, the method comprising: S100, acquiring remote sensing images of the target water body covering the visible light to thermal infrared bands in the first and second time phases within a set evaluation period, as well as synchronous meteorological data, and performing correction and registration processing to obtain the remote sensing reflectance of the water body and matching wind speed and atmospheric carbon dioxide concentration data; S200, retrieving the spatial distribution of dissolved organic carbon and particulate organic carbon concentrations in the water body based on the remote sensing reflectance of the first and second time phases respectively, and performing vertical integration calculations in conjunction with the water depth data of the target water body to obtain the total carbon storage per unit area of ​​water column in the corresponding time phase; S300, retrieving the water surface temperature and chlorophyll a concentration based on the remote sensing images respectively, and using an estimation model to determine the carbon dioxide concentration in the water surface layer. Carbon partial pressure is calculated by determining the carbon dioxide partial pressure difference at the water-air interface between the water body and the atmosphere, and the gas transport rate is calculated based on the wind speed. The carbon dioxide exchange flux at the water-air interface is obtained by combining the carbon dioxide solubility constant and the carbon dioxide partial pressure difference at the water-air interface. S400: The carbon dioxide exchange flux is accumulated in time and space within a set evaluation period. The change in total carbon storage per unit area of ​​water column and the amount of exogenous carbon input are calculated by subtracting the total carbon storage of the second phase from that of the first phase within the same time period. The flow rate and river carbon concentration of the inflow section of the target water body are obtained to calculate the amount of exogenous carbon input. A collaborative accounting equation is established for coupled calculation. After deducting the contribution of external input, the endogenous net carbon sink generated by the biogeochemical processes inside the water body is output to determine the net carbon source and sink attributes of the water body.

[0007] Optionally, in step S200, different bio-optical inversion paths are adaptively selected based on a preset water turbidity discrimination threshold; wherein, for turbid water, an inversion model based on a combination index is used to invert the dissolved organic carbon concentration, and the inversion process is expressed by the following formula:

[0008]

[0009]

[0010] Indicates the portfolio index. , and These represent the remote sensing reflectance of the first, second, and third bands, respectively. Indicates the concentration of dissolved organic carbon. Represents the regression coefficient. Indicates the intercept;

[0011] When retrieving particulate organic carbon concentration, a normalized difference index is constructed and the inversion is performed using an exponential function. The calculation process is expressed by the following formula:

[0012]

[0013]

[0014] in Indices the normalized difference index of particulate organic carbon and These represent the remote sensing reflectance of the fourth and fifth bands, respectively. Indicates particulate organic carbon concentration. and This represents the model constant coefficients determined through calibration using measured data.

[0015] Optionally, in step S200, when obtaining the total carbon storage per unit area of ​​water column, it is assumed that the carbon component concentration decreases exponentially with increasing water depth, and the stratified vertical integral calculation process of the total carbon storage per unit area of ​​water column is expressed by the following formula:

[0016]

[0017]

[0018]

[0019] in Indicates passing through the sixth band and the seventh band The effective water depth is estimated from the relationship between remote sensing reflectance. and Indicates calibration parameters, Indicates water depth The concentration of carbon components at that location, This represents the concentration of dissolved organic carbon or particulate organic carbon in the surface layer obtained from remote sensing inversion. This represents the diffuse attenuation coefficient estimated from the slope of the remotely sensed reflectance spectrum. This represents the total carbon storage per unit area of ​​water column. This indicates the total number of layers after rounding up the effective water depth. and They represent the first The dissolved organic carbon concentration and particulate organic carbon concentration were calculated using the exponential decay law. Indicates the first The thickness of the layer.

[0020] Optionally, in step S300, the water surface temperature is retrieved using a split-window algorithm and the chlorophyll a concentration is retrieved using a three-band model, specifically including: retrieving the water surface temperature using a split-window algorithm based on the brightness temperature of adjacent thermal infrared channels, and expressing the calculation process of the water surface temperature using the following formula:

[0021]

[0022] The chlorophyll a concentration was retrieved using a three-band model based on the separation of fluorescence signals from red and near-infrared reflectance, and the calculation process for extracting the fluorescence peak height was expressed by the following formula:

[0023]

[0024] in, This represents the water surface temperature obtained through inversion. and These represent the brightness temperatures obtained after radiometric calibration and atmospheric correction for the first and second thermal infrared channels, respectively. and Indicates the atmospheric water vapor correction factor;

[0025] Furthermore, the process of obtaining the atmospheric water vapor correction coefficient includes:

[0026] Using the radiative transfer model, the atmospheric profile parameters during transit are input, and the atmospheric transmittance of the first thermal infrared channel and the second thermal infrared channel is calculated and output.

[0027] Based on the atmospheric transmittance, interpolation is performed in a pre-established lookup table to obtain the corresponding atmospheric water vapor correction coefficient. and ;

[0028] This indicates the height of the fluorescence peak located near the ninth band. These represent the remote sensing reflectance of the eighth, ninth, and tenth bands, respectively.

[0029] Optionally, in step S300, the estimation model for estimating the partial pressure of carbon dioxide in the water surface layer is a nonlinear multiple regression model, and its calculation process is expressed by the following formula:

[0030]

[0031] in, This indicates the partial pressure of carbon dioxide in the surface layer of water. This indicates the inverted water surface temperature. This represents the inverted chlorophyll a concentration. This represents the regression coefficients of the model determined by fitting the training dataset.

[0032] Optionally, in step S300, the calculation process of carbon dioxide exchange flux at the water-air interface is expressed by the following formula:

[0033]

[0034] in, This represents the carbon dioxide exchange flux. This represents the gas transport rate calculated based on wind speed and Schmidt number. This represents the solubility constant of carbon dioxide in water. This represents the difference in carbon dioxide partial pressure at the water-air interface calculated from the partial pressure of carbon dioxide in water and the partial pressure of carbon dioxide in the atmosphere.

[0035] Optionally, when applied to the calculation of water bodies during high-latitude ice-covered periods, the method further includes an adaptive adjustment step for ice-covered periods: using synthetic aperture radar data to assist in determining the water body boundary; simplifying the carbon dioxide partial pressure estimation model of the water surface by setting the chlorophyll a concentration-related terms to zero, so that the carbon dioxide partial pressure estimation model retains only terms related to water surface temperature; in step S300, an ice cover factor is introduced to characterize the inhibitory effect of the ice cover on water-air interface exchange, and the calculation process of the corrected gas transport rate is expressed by the following formula:

[0036]

[0037] in, This indicates the corrected gas transport rate. The constant proportionality coefficient is determined through calibration. This indicates the wind speed at a height of 10 meters. The Schmidt number represents the relationship between water temperature and temperature. The value of the ice cover factor is 0 to 1, where 0 indicates no ice cover and 1 indicates complete freezing.

[0038] Optionally, in step S400, the collaborative accounting equation for deducting external input contributions is expressed by the following formula:

[0039]

[0040] in, This indicates the amount of endogenous net carbon sequestration. This represents the cumulative carbon dioxide exchange flux within the set evaluation period. This indicates the change in the total carbon storage of the water column within the same time period. This indicates the amount of exogenous carbon input from outside the watershed into the target water area, and all parameters are uniformly converted to the same unit of carbon molarity.

[0041] Furthermore, the calculation process for the amount of exogenous carbon input is expressed by the following formula:

[0042]

[0043] in, This represents the total amount of exogenous carbon input obtained per unit area of ​​the target water body. Indicates the first Indicates the first River carbon concentration at the main stream cross section This represents the total time of the evaluation time window. This indicates the total area of ​​the target water body.

[0044] Optionally, the method further includes: S500, applying random perturbations conforming to their typical error distributions to key input parameters such as remote sensing reflectance, water depth, wind speed, and atmospheric carbon dioxide concentration using the Monte Carlo simulation method; S600, obtaining the probability distribution statistical characteristics of endogenous net carbon sink by repeatedly executing the complete accounting process, thereby quantifying and outputting the uncertainty of the final assessment result.

[0045] Optionally, when applied to the accounting of small bodies of water, the method includes the following adaptive optimization process:

[0046] Remote sensing data acquired using an unmanned aerial vehicle (UAV) platform was used as the primary data source, and an adapted atmospheric correction model was employed.

[0047] High-precision water depth data is obtained using airborne lidar technology;

[0048] A model for estimating the partial pressure of carbon dioxide in water is obtained by training localized samples established for the target water body; and the evaluation time window is shortened accordingly to capture carbon dynamic changes over shorter periods.

[0049] The beneficial effects of this application are:

[0050] This invention acquires multi-source remote sensing images and synchronous meteorological data, and uses remote sensing reflectance to invert water organic carbon concentration, water surface temperature, and chlorophyll a concentration. It solves the problems of traditional methods that rely on in-situ sampling and are limited by manpower, material resources, and extreme weather. It enables dynamic monitoring of river and lake carbon sinks on a large scale, high frequency, and continuous spatial scale, and significantly improves monitoring efficiency and coverage.

[0051] Meanwhile, this invention calculates the dynamic change of total carbon storage in the water column by introducing a difference mechanism between remote sensing data of the first and second time phases, and constructs a collaborative accounting equation by combining the spatiotemporal accumulation of carbon dioxide exchange flux at the water-air interface. This overcomes the defect of separate assessment of static carbon storage and dynamic flux in the prior art, and realizes the systematic coupling and rigorous accounting of carbon sink process in water bodies.

[0052] Meanwhile, by obtaining the flow rate and river carbon concentration at the inflow section of the main stream, this invention quantitatively calculates the amount of exogenous carbon input and deducts it in the collaborative accounting equation. This solves the problem that the existing accounting system ignores the interference of exogenous carbon input in the watershed. It can accurately output the endogenous net carbon sink that reflects the biogeochemical processes inside the water body, and significantly improves the accuracy and reliability of the determination of the net carbon source and sink attributes of water bodies.

[0053] Furthermore, this invention designs an adaptive adjustment step for ice-covered water bodies and an adaptive optimization process for small water bodies, specifically for special scenarios such as high-latitude ice-covered water bodies and small water bodies. By introducing ice cover factors, simplifying the estimation model, and employing technologies such as UAVs and lidar, the universality and calculation accuracy of the method under different environmental conditions are enhanced.

[0054] Finally, this invention uses the Monte Carlo simulation method to apply random perturbations to key input parameters and repeat the calculation process multiple times to obtain the probability distribution statistical characteristics of endogenous net carbon sinks, quantify the uncertainty of the assessment results, and provide rigorous statistical confidence and engineering reference value for carbon sink assessment. Attached Figure Description

[0055] Figure 1 This is an overall flowchart of a river and lake carbon sequestration accounting method based on remote sensing data provided in an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram illustrating the principle of remote sensing inversion for a river and lake carbon sequestration accounting method based on remote sensing data, provided in an embodiment of the present invention.

[0057] Figure 3 This is a schematic diagram of time series accounting based on remote sensing inversion, which is a method for calculating carbon sinks in rivers and lakes based on remote sensing data, provided in an embodiment of the present invention.

[0058] Figure 4 This is a schematic diagram illustrating the process of transforming an original image into a carbon distribution map, provided in an embodiment of the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0060] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0061] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0062] As mentioned earlier, existing methods for monitoring and accounting for carbon sinks in rivers and lakes have significant shortcomings. On the one hand, traditional methods rely heavily on in-situ contact sampling, which is limited by manpower, resources, and extreme weather, making it impossible to achieve large-scale, high-frequency, and continuous spatial-scale dynamic monitoring. On the other hand, existing technologies often isolate the carbon sink process in water bodies, assessing only the static carbon storage or the instantaneous carbon dioxide exchange flux at the water-air interface in isolation, making it difficult to systematically couple the spatiotemporal integral of dynamic flux with changes in carbon storage. More critically, existing accounting systems typically ignore the severe interference from external carbon inputs to the watershed, resulting in calculations that fail to accurately reflect the endogenous net carbon sink generated by biogeochemical processes within the water body, leading to significant biases in the final determination of net carbon source and sink attributes.

[0063] To address this issue, the present invention provides a method for calculating carbon sequestration in rivers and lakes based on remote sensing data, which solves the above problems. The present invention solves the problem in the following way.

[0064] Example 1:

[0065] Please refer to the instruction manual appendix. Figures 1 to 4 As shown in the figure, this embodiment provides a method for calculating river and lake carbon sequestration based on remote sensing data.

[0066] The method includes the following steps:

[0067] The first step, S100, involves multi-source data acquisition and preprocessing. This includes acquiring raw remote sensing images covering the visible to thermal infrared bands of the target water area, along with synchronized meteorological data. Atmospheric correction is applied to the remote sensing images (preferably using the 6S radiative transfer model or the ACOLITE algorithm suitable for water color remote sensing to remove atmospheric path radiation interference) to obtain the remote sensing reflectance of the water body. The meteorological data is then standardized and spatiotemporally registered to obtain wind speed, atmospheric carbon dioxide concentration, and air pressure data that match the remote sensing observations. This step aims to provide high-precision, spatiotemporally aligned foundational data for subsequent carbon sequestration accounting.

[0068] In this embodiment, step S200 further includes: adaptively selecting different bio-optical inversion paths based on a preset water turbidity discrimination threshold. This mechanism aims to adapt to the optical characteristics of water bodies with different turbidity levels. For turbid water bodies, an inversion model based on a combination index is used to invert the dissolved organic carbon concentration, and the inversion process is expressed by the following formula:

[0069]

[0070]

[0071] in, This represents the combined index. , and These correspond to the near-ultraviolet to blue light bands where dissolved organic carbon in water has significant absorption characteristics. When performing carbon sink accounting for different rivers and lakes, those skilled in the art can select the channels with center wavelengths closest to 380nm, 412nm, and 443nm as the first, second, and third bands, respectively, based on the remote sensing sensor used. , and These represent the corresponding remote sensing reflectances. This indicates the concentration of dissolved organic carbon. Represents the regression coefficient. The intercept is represented by the regression coefficient; where the regression coefficient is... With intercept The following general methods are used for calibration and acquisition:

[0072] Constructing a calibration sample set: Collecting ground-measured dissolved organic carbon from the target water area. Concentration data is used as the dependent variable. Remote sensing images that are spatiotemporally matched with the measured data are acquired simultaneously. The remote sensing reflectance of the corresponding points is extracted and the combined index is calculated. As independent variables; among them, the spatiotemporal matching requirements are: the difference between the measured sampling time and the remote sensing image transit time shall not exceed 24 hours, and the spatial distance between the measured point and the remote sensing pixel center shall not exceed 1 pixel resolution; the effective sample size of the sample set shall not be less than 30 sets, and it shall cover different seasons and different hydrological conditions of the target water area;

[0073] Data preprocessing: Outlier removal was performed on the sample set. The Grubbs test was used to remove outlier samples at a 95% confidence level. The remaining samples were used for fitting.

[0074] Model Fitting: A univariate linear regression was performed using the least squares method to obtain the regression coefficients. With intercept After fitting, the model is validated, requiring a coefficient of determination R² ≥ 0.6 and a root mean square error RMSE ≤ 1.5 mg / L to ensure that the model accuracy meets the calculation requirements.

[0075] When retrieving particulate organic carbon concentration, a normalized difference index is constructed and the inversion is performed using an exponential function. The calculation process is expressed by the following formula:

[0076]

[0077]

[0078] in, The normalized difference index for particulate organic carbon is used. In practical applications targeting different rivers and lakes, sensor channels with center wavelengths closest to 560nm and 665nm can be selected as the fourth and fifth bands, respectively. and These represent the remote sensing reflectance of the fourth and fifth bands, respectively. Indicates particulate organic carbon concentration. and This represents the model constant coefficients determined through calibration using measured data. They are obtained by using measured particulate organic carbon concentration data from the target water area and simultaneously extracted... The sample set is constructed using an exponential function and then fitted and calibrated using an exponential function or a nonlinear least squares method.

[0079] The model constant coefficients , The following general methods are used for calibration and acquisition:

[0080] Constructing a calibration sample set: Collecting ground-based measured particulate organic carbon data from the target water area. Using concentration data as the dependent variable, remote sensing images that are spatiotemporally matched with the measured data are acquired simultaneously. The remote sensing reflectance of the corresponding points is extracted, and the normalized difference index of particulate organic carbon is calculated. As independent variables; the spatiotemporal matching requirements and sample size requirements are the same as described above. Model;

[0081] Data preprocessing: Same as above Outlier removal rules for the model;

[0082] Model Fitting: The exponential function model was fitted using the nonlinear least squares method, with an iterative convergence threshold set to 10. -6 The maximum number of iterations was set to 1000, and constant coefficients c1 and c2 were obtained through fitting. After fitting, the coefficient of determination R was verified. 2 ≥0.55, root mean square error RMSE ≤1.2mg / L.

[0083] In this first embodiment, step S200 further includes: when obtaining the total carbon storage per unit area of ​​water column, assuming that the carbon component concentration decreases exponentially with increasing water depth, the surface concentration is extended to the entire water column, and the stratified vertical integral calculation process of the total carbon storage per unit area of ​​water column is expressed by the following formula:

[0084]

[0085]

[0086]

[0087] in This indicates the estimated effective water depth. and These correspond to the shortwave infrared bands that are strongly affected by water absorption. When applying to different water bodies, the channels with center wavelengths closest to 1640nm and 2130nm can be selected as the sixth and seventh bands, respectively. and These represent the corresponding remote sensing reflectances. and Indicates calibration parameters,

[0088] The calibration parameters , The following general methods are used for calibration and acquisition:

[0089] Constructing a calibration sample set: Collect measured effective water depth data of the target water area (acquired by shipborne sonar / airborne lidar bathymetry) as the dependent variable, simultaneously acquire matching remote sensing images, and calculate... As an independent variable; the spatiotemporal matching requirements are that the difference between the depth measurement time and the image transit time should not exceed 72 hours, the sample size should not be less than 20 groups, and the samples should cover different water depth gradients.

[0090] Data preprocessing: Samples from shallow water areas with a water depth of <0.5m and areas covered by aquatic vegetation were removed, and outliers were removed using the Grubbs test.

[0091] Model fitting: A univariate linear regression was performed using the least squares method to obtain the model. (Regression slope) and (Intercept); Verification requires coefficient of determination R. 2 ≥0.7, RMSE≤0.5m.

[0092] Indicates water depth The concentration of carbon components at that location, This represents the concentration of dissolved organic carbon or particulate organic carbon in the surface layer obtained from remote sensing inversion. This represents the diffuse attenuation coefficient estimated by the spectral slope of remotely sensed reflectance. It is obtained by establishing an empirical regression model based on the measured diffuse attenuation coefficient of the target water area and the synchronously sensed reflectance (preferably the ratio of the green band to the red band or the spectral slope). The empirical regression model is a power function model, and its expression is: ,in and These represent the corresponding remote sensing reflectance, parameters. and The diffuse attenuation coefficient was determined by fitting the sample set of measured diffuse attenuation coefficients in the target water area using the least squares method.

[0093] This represents the total carbon storage per unit area of ​​water column. This indicates the total number of layers after rounding up the effective water depth. and They represent the first The dissolved organic carbon concentration and particulate organic carbon concentration were calculated using the exponential decay law. Indicates the first The thickness of the layer.

[0094] In this first embodiment, step S300 obtains water temperature through thermal infrared remote sensing and combines it with chlorophyll a concentration, which reflects biological activity, to drive the estimation model. The water surface temperature is retrieved using a split-window algorithm, and the chlorophyll a concentration is retrieved using a three-band model. The water surface temperature is retrieved using a split-window algorithm based on the brightness temperature of adjacent thermal infrared channels, and the calculation process of the water surface temperature is expressed by the following formula:

[0095]

[0096] The chlorophyll a concentration was retrieved using a three-band model based on the separation of fluorescence signals from red and near-infrared reflectance, and the calculation process for extracting the fluorescence peak height was expressed by the following formula:

[0097]

[0098] in, This represents the water surface temperature obtained through inversion. and These represent the brightness temperatures obtained after radiometric calibration and atmospheric correction for the first and second thermal infrared channels, respectively. In practical applications, those skilled in the art can select two adjacent thermal infrared bands with center wavelengths around 10.8 μm (corresponding to the first thermal infrared channel) and around 12.0 μm (corresponding to the second thermal infrared channel), respectively, based on the remote sensing sensor used. and Indicates the atmospheric water vapor correction factor;

[0099] Furthermore, the process of obtaining the atmospheric water vapor correction coefficient includes:

[0100] Using the radiative transfer model, the atmospheric profile parameters during transit are input, and the atmospheric transmittance of the first thermal infrared channel and the second thermal infrared channel is calculated and output.

[0101] Based on the atmospheric transmittance, interpolation is performed in a pre-established lookup table to obtain the corresponding atmospheric water vapor correction coefficient. and The pre-established lookup table is obtained by forward simulation calculation of atmospheric transmittance and correction coefficients using the MODTRAN atmospheric radiative transfer model under different atmospheric water vapor content gradients and surface temperature gradients. , The corresponding datasets between them are established.

[0102] This indicates the height of the fluorescence peak located near the ninth band. , and These correspond to the red light absorption baseline band, fluorescence peak band, and near-infrared baseline band required for extracting chlorophyll fluorescence signals, respectively. In practical applications, preferably, the channel with the center wavelength closest to 665nm can be selected as the eighth band, the channel with the center wavelength closest to 705nm as the ninth band, and the channel with the center wavelength closest to 740nm as the tenth band. and These represent their corresponding remote sensing reflectances.

[0103] Obtaining fluorescence peak height Then, the chlorophyll a concentration in the water body was calculated using a univariate linear empirical model. The calculation process is expressed by the following formula:

[0104]

[0105] in, This represents the inverted chlorophyll a concentration. and These represent the transformation slope and intercept, respectively. The transformation coefficients... and Obtained through calibration using the following general method:

[0106] Measured chlorophyll a concentration data of the target water area were collected as the dependent variable, and matching remote sensing images were acquired simultaneously to calculate the fluorescence peak height at the corresponding locations. As the independent variable; after constructing a sample set and removing outliers, the least squares method is used for linear regression fitting to determine the conversion slope suitable for the target water area. and intercept .

[0107] In this embodiment, step S300 further includes: the estimation model for estimating the partial pressure of carbon dioxide on the water surface is a nonlinear multiple regression model. This model explicitly includes the control of temperature on physical solubility and the nonlinear influence of chlorophyll on biological activity. Its calculation process is expressed by the following formula:

[0108]

[0109] in, This indicates the partial pressure of carbon dioxide in the surface layer of water. This indicates the inverted water surface temperature. This represents the inverted chlorophyll a concentration. This represents the regression coefficients of the model determined by fitting the training dataset.

[0110] The model regression coefficients to It is obtained through training using the following general method:

[0111] Construct a training dataset: Collect in-situ measured surface carbon dioxide partial pressure (pCO2) data of the target water body as the dependent variable, and simultaneously acquire spatiotemporally matched surface temperature (SST) and chlorophyll a concentration (Chl-a) data as independent variables; the spatiotemporal matching requirement is that the difference between the sampling time and the environmental data acquisition time does not exceed 2 hours, the effective sample size is not less than 50 groups, and the data covers different seasons, water temperatures and nutrient levels.

[0112] Dataset partitioning: Randomly divide the dataset into training and validation sets in a 7:3 ratio;

[0113] Model Fitting and Validation: A multivariate nonlinear least squares method was used for fitting, with an iterative convergence threshold set to 10. -6 To obtain the regression coefficients to The validation set requires a coefficient of determination R. 2 ≥0.65, mean relative error (MRE) ≤15%.

[0114] Based on the derived partial pressure difference and wind speed environmental dynamic parameters, the final flux is obtained using thin-film theory. The calculation process for the carbon dioxide exchange flux at the water-air interface is expressed by the following formula:

[0115]

[0116] in, This represents the carbon dioxide exchange flux. This represents the gas transport rate calculated based on wind speed and Schmidt number. This represents the solubility constant of carbon dioxide in water.

[0117] Its value is based on the conventional values ​​in this field. An empirical formula is used to obtain the water surface temperature. The target water salinity is substituted into the calculation; for inland freshwater rivers and lakes, the water salinity is approximately taken as 0, or measured salinity data is obtained through a synchronous automatic water quality monitoring station. This represents the gas transport rate, and its value is calculated based on the ambient wind speed and the Schmidt number related to water temperature (e.g., using...). Empirical models of conventional gas transport rates in this field, in which The constant proportionality coefficient, the constant proportionality coefficient The calibration method is as follows: deploy in-situ observation equipment for water-air interface flux in the target water area, simultaneously acquire measured gas transport rate and corresponding wind speed and water temperature data, construct a calibration sample set, and then use nonlinear least squares fitting to determine the calibration. This represents the difference in carbon dioxide partial pressure at the water-air interface calculated from the partial pressure of carbon dioxide in water and the partial pressure of carbon dioxide in the atmosphere.

[0118] In this first embodiment, for all the semi-empirical or empirical model parameters mentioned above (including linear regression coefficients) and intercept constant coefficients and Water depth calibration parameters and And the model regression coefficients for calculating carbon dioxide partial pressure. to The universal calibration process is as follows: Historical ground station data from the same period in the target water area to be calibrated are collected in advance (specifically including dissolved organic carbon concentration, particulate organic carbon concentration, measured water depth, water surface temperature, and chlorophyll a concentration, etc.), and synchronous remote sensing spectral feature data (such as reflectance of each band and its combination index) are extracted from the corresponding locations to construct a training dataset. For linear relationships (such as...), the calibration process is as follows: , , , Based on the training dataset, the least squares method is used for conventional linear regression fitting and calibration; for nonlinear relationships (such as...), ... , , to The nonlinear least squares method or a combination of machine learning algorithms such as random forest is used to perform multivariate fitting and calibration, thereby obtaining localization coefficient values ​​that are adapted to the characteristics of the target water area.

[0119] Regarding the preset water turbidity discrimination threshold, this embodiment uses a measured suspended solids concentration equal to 10 mg / L or a remote sensing reflectance in the infrared band (e.g., the eighth band) greater than a preset baseline reflectance threshold (e.g., ...). () as a classification criterion.

[0120] In this embodiment, step S400 further includes: establishing a collaborative accounting equation and outputting the endogenous net carbon sink. The carbon dioxide exchange flux is accumulated temporally and spatially within a set evaluation period (specifically: the carbon dioxide exchange flux of all effective remote sensing pixels in the target water area is spatially averaged and then multiplied by the total time of the evaluation window to obtain the cumulative carbon dioxide exchange flux representing the unit area of ​​the target water area). The total carbon storage in the second time phase and the first time phase obtained in step S200 are subtracted to calculate the change in the total carbon storage per unit area of ​​the water column within the same time period (specifically, the spatial average of the total carbon storage of all effective remote sensing pixels in the target water area under each time phase is calculated, and the change is obtained by subtracting the average of the first time phase from the average of the second time phase). The dynamic water surface width of each inflow section is extracted using radar satellite imagery, and the water surface elevation data obtained by altimetry satellites is combined with the inversion of the water level-flow relationship curve to obtain the flow rate of the inflow section. The synchronous total organic carbon (i.e., the sum of dissolved organic carbon and particulate organic carbon) concentration of the river is obtained through automatic water quality monitoring station data of the inflow section of the target water area, conventional manual sampling measurement data, or mature online monitoring methods such as nondispersive infrared absorption spectroscopy (NDIR) to calculate the exogenous carbon input. To eliminate the interference of endogenous processes in non-target water areas, it is necessary to estimate the exogenous carbon input from outside the watershed, and the calculation process of the exogenous carbon input is expressed by the following formula:

[0121]

[0122] in, This represents the total amount of exogenous carbon input obtained per unit area of ​​the target water body. Indicates the first Indicates the first River carbon concentration at the main stream cross section This represents the total time of the evaluation time window. This indicates the total area of ​​the target water body.

[0123] Subsequently, a collaborative accounting equation is established for coupled computation. The collaborative accounting equation used to deduct external input contributions is expressed by the following formula:

[0124]

[0125] in, This indicates the amount of endogenous net carbon sequestration. This represents the cumulative carbon dioxide exchange flux within the set evaluation period. This indicates the change in the total carbon storage of the water column within the same time period. This indicates the amount of exogenous carbon input from outside the watershed into the target water area, and all parameters are uniformly converted to the same carbon molar unit.

[0126] Example 2:

[0127] Based on the above embodiments, in order to further explain in detail the specific operation process of the accounting method in Embodiment 1 under different application scenarios, this embodiment provides an implementation method with specific parameters, which integrates the adaptation optimization of water bodies under different scales and different climatic conditions.

[0128] The specific values ​​of model coefficients, band wavelengths, meteorological parameters, etc., substituted in this embodiment are exemplary data pre-fitted and calibrated based on local measured sample sets for this specific inland complex lake simulation scenario. They are only used to clearly illustrate the multi-step coupled calculation logic and accounting process of the present invention and do not constitute any numerical limitation on the scope of protection of the present invention. When performing carbon sink accounting for other river and lake water bodies, those skilled in the art can determine the parameter values ​​suitable for the target water area according to the general parameter selection rules and calibration methods disclosed above in this specification.

[0129] This embodiment simulates a carbon sequestration calculation for a typical inland composite lake. The target water area includes a large main lake area with an area greater than 1 square kilometer and a small urban landscape lake area with an area less than 1 square kilometer, and the water area experiences a significant ice-covered period in winter. The simulation is conducted under clear, cloudless summer weather conditions, with a 7-day evaluation time window. It should be noted that the model coefficients (such as...) substituted into the subsequent calculations in this embodiment... , , The specific values ​​of parameters such as meteorological parameters are exemplary data that are pre-fitted and calibrated or assumed based on local measured sample sets for the specific lake simulation scenario. They are only used to clearly illustrate the multi-step coupled calculation logic and accounting process of the present invention and do not constitute any numerical limitation on the scope of protection of the present invention.

[0130] The specific implementation process is broken down as follows:

[0131] Step S100: Data Acquisition and Processing. Acquire remote sensing images and meteorological data of the target water area. For large main lake areas, acquire Level-1 remote sensing data from the Landsat 8 satellite mounted on a polar-orbiting platform. The sensors include a multispectral imager and a TIRS sensor, with a spatial resolution between 30m and 1000m. For small scenic lake areas, use the DJI M300RTK drone platform, equipped with a Headwall Nano-Hyperspec VNIR hyperspectral sensor (spatial resolution approximately 1.5m, ground sampling distance approximately 1cm), a Riegl mini VUX-2 UAV airborne lidar, and a TeAx Thermo IMAGER lightweight thermal infrared camera, with the flight altitude set at 120m.

[0132] If the monitoring occurs during a high-latitude ice-covered period, Sentinel-1C band synthetic aperture radar (SAR) data with all-weather cloud penetration capability will be prioritized. Meteorological data are sourced from the lake's micro-weather station and the surrounding network of automatic weather stations. Wind speed data will be uniformly interpolated to a height of 10m and denoted as [data missing]. .

[0133] Step S200: Static carbon storage calculation

[0134] Based on the preset turbidity threshold, the area in Example 2 is classified as turbid water. The dissolved organic carbon concentration is retrieved using an inversion model based on a combination index, and the inversion process is expressed by the following formula:

[0135]

[0136]

[0137] Substitute the measured parameters for the current point: Let the first band be... (380nm), second band (412nm) and the third band Remote sensing reflectance Rrs(443nm) ),Rrs( ) and Rrs( The values ​​are 0.045, 0.025, and 0.015 per steradian, respectively. Calculate the combination index. .

[0138] in, Indicates the portfolio index. , and These represent the remote sensing reflectance of the first, second, and third bands, respectively. Represents the concentration of dissolved organic carbon, regression coefficient The value is 450.0, and the intercept is... If the value is 2.5, then =450.0×0.005+2.5=4.75mg / L.

[0139] When retrieving particulate organic carbon concentration, a normalized difference index is constructed and the inversion is performed using an exponential function. The calculation process is expressed by the following formula:

[0140]

[0141]

[0142] Substitute the measured parameters for the current point: Let's assume the fourth band... With the fifth band Remote sensing reflectance and These are 0.032 and 0.005 per steradian, respectively. The calculation yields...

[0143]

[0144] in, This represents the normalized difference index of particulate organic carbon. and These represent the remote sensing reflectance of the fourth and fifth bands, respectively. Represents particulate organic carbon concentration, model constant coefficients The value is 18.7. If the value is -3.2, then mg / L.

[0145] When obtaining the total carbon storage per unit area of ​​water column, it is assumed that the carbon concentration decreases exponentially with increasing water depth, and the stratified vertical integral calculation process of the total carbon storage per unit area of ​​water column is expressed by the following formula:

[0146]

[0147]

[0148]

[0149] Set the sixth band (1640nm) and the seventh band Remote sensing reflectance at (2130nm) and The calibration parameters are 0.002 and 0.001 per steradian, respectively. and The values ​​are 2.5 and 1.0 respectively. The calculated effective water depth is... m.

[0150] Regarding the diffuse attenuation coefficient Using the aforementioned power function model, the fourth band is extracted. (i.e., the green light band, 560nm) and the fifth band Remote sensing reflectance (i.e., red light band, 665nm) and Set the model parameters calibrated for the target lake. , Calculation ; to water layer The thickness m is divided into N=3 layers. For example, If the surface concentration is 1.81, then the concentration of the first layer (z=0.5) is... mg / L.

[0151] Similarly, calculate each layer and After concentration, multiply by thickness By accumulating the values ​​layer by layer, the total carbon storage per unit area of ​​water column was finally obtained. mg / cm 2 .

[0152] in, Indicates the effective water depth. and Indicates calibration parameters, Indicates water depth The concentration of carbon components at that location, This represents the concentration of dissolved organic carbon or particulate organic carbon in the surface layer obtained from remote sensing inversion. This represents the diffuse attenuation coefficient estimated from the slope of the remotely sensed reflectance spectrum. This represents the total carbon storage per unit area of ​​water column. This indicates the total number of layers after rounding up the effective water depth. and They represent the first The dissolved organic carbon concentration and particulate organic carbon concentration were calculated using the exponential decay law. Indicates the first The thickness of the layer.

[0153] Step S300: Dynamic carbon flux accounting

[0154] The water surface temperature is retrieved using a split-window algorithm based on the brightness temperature of adjacent thermal infrared channels, and the calculation process is expressed by the following formula:

[0155]

[0156] Substitute the current satellite data: .in, This represents the water surface temperature obtained through inversion. and These represent the brightness temperatures obtained after radiometric calibration and atmospheric correction for the first and second thermal infrared channels, respectively (corresponding to band 10 298K and band 11 297K of the Landsat 8 satellite in this embodiment). and This represents the atmospheric water vapor correction factor, with values ​​of 1.0 and 0 respectively.

[0157] The chlorophyll a concentration was retrieved using a three-band model based on the separation of fluorescence signals from red and near-infrared reflectance, and the calculation process for extracting the fluorescence peak height was expressed by the following formula:

[0158]

[0159] Substitute parameters for calculation: Assume the eighth band Band 9 and the tenth band The remote sensing reflectivities were 0.005, 0.012, and 0.002, respectively.

[0160] Calculate fluorescence peak height Set a pre-calibrated conversion slope for the target lake. The value is 2500, and the intercept is... The value is set to 1.5. Substituting this value into the chlorophyll a concentration conversion formula, the inverted chlorophyll a concentration is obtained. .

[0161] in, This indicates the height of the fluorescence peak located near the ninth band. , and These represent the remote sensing reflectance of the eighth, ninth, and tenth bands, respectively.

[0162] The estimation model for the partial pressure of carbon dioxide in the water surface is a nonlinear multiple regression model, and its calculation process is expressed by the following formula:

[0163]

[0164] The calculated and Substituting the values ​​into the model (with regression coefficients α0 to α5 pre-trained), the partial pressure of carbon dioxide in the water surface is obtained. .

[0165] At this time, the atmospheric carbon dioxide partial pressure measured simultaneously was 410. The partial pressure difference of carbon dioxide at the water-air interface was calculated. .

[0166] in, This indicates the partial pressure of carbon dioxide in the surface layer of water. This indicates the inverted water surface temperature. This represents the inverted chlorophyll a concentration. to This represents the regression coefficients of the model determined by fitting the training dataset.

[0167] The calculation process for carbon dioxide exchange flux at the water-air interface is expressed by the following formula:

[0168]

[0169] Substituting the above parameters: Let the gas transport rate be calculated based on the ambient wind speed. The solubility constant of carbon dioxide in water The carbon dioxide exchange flux was calculated. .

[0170] in, This represents the carbon dioxide exchange flux. This represents the gas transport rate calculated based on wind speed and Schmidt number. This represents the solubility constant of carbon dioxide in water. This represents the difference in carbon dioxide partial pressure at the water-air interface calculated from the partial pressure of carbon dioxide in water and the partial pressure of carbon dioxide in the atmosphere.

[0171] Step S400, Collaborative Analysis and Endogenous Net Carbon Sequestration Output

[0172] The calculation process for exogenous carbon input is expressed by the following formula:

[0173]

[0174] Substituting hydrological parameters: The discharge of a certain main stream is obtained by coupling radar satellite and altimetry data. River carbon concentration The total evaluation time window is Δt = 604800s (7 days), and the total area of ​​the target water body is... ㎡. The total input quantity is calculated and uniformly converted to obtain... .

[0175] in, This represents the total amount of exogenous carbon input obtained per unit area of ​​the target water body. Indicates the first The flow rate at the inflow section of the main stream, Indicates the first The river carbon concentration at the main stream cross-section, where Δt represents the total time of the evaluation time window. This indicates the total area of ​​the target water body.

[0176] The collaborative accounting equation used to deduct contributions from external inputs is expressed as follows:

[0177]

[0178] Substitute the cumulative parameters for this period: the cumulative carbon dioxide exchange flux after spatial averaging across the entire lake within the set evaluation period (7 days). The change in total carbon storage in the water column within the same time period is calculated by subtracting the spatial average of total carbon storage obtained from the second time phase and the first time phase. Then the endogenous net carbon sink .

[0179] in This indicates the amount of endogenous net carbon sequestration. This represents the cumulative carbon dioxide exchange flux within the set evaluation period. This indicates the change in the total carbon storage of the water column within the same time period. This indicates the amount of exogenous carbon input from outside the watershed into the target water area, and all parameters are uniformly converted to the same carbon molar unit.

[0180] Based on calculations A negative value indicates that the water body has net carbon sink properties during the set evaluation period.

[0181] Example 3:

[0182] Based on the above embodiments, in order to further improve the universality and reliability of the present invention in complex and special aquatic environments, this embodiment, based on Embodiment 1 and Embodiment 2, details the adaptive optimization mechanism for special aquatic environments, and fully provides the specific implementation process of uncertainty quantification and result evaluation in steps S500 and S600.

[0183] In actual calculations, when the target water area possesses specific environmental attributes, the system will trigger the following adaptive optimization process during the execution of steps S100 to S400:

[0184] Firstly, when applied to the calculation of small water bodies, the method includes the following adaptive optimization process: in the data acquisition stage, remote sensing data is acquired using an unmanned aerial vehicle (UAV) platform as the main data source, and an adapted atmospheric correction model is adopted (e.g., empirical linear correction based on known reflectivity grayscale targets deployed synchronously on the ground, or an empirical radiative transfer model specifically for UAVs); in the water depth data acquisition stage, high-precision water depth data is acquired using airborne lidar technology.

[0185] In the dynamic flux accounting stage, due to the significant environmental disturbances affecting small water bodies, it is necessary to use localized samples established for the target water body to train and obtain a water body carbon dioxide partial pressure estimation model; and correspondingly shorten the evaluation time window to capture shorter-cycle carbon dynamic changes. Secondly, when applied to high-latitude water bodies during the ice-covered period for accounting, the method also includes an ice-covered period adaptive adjustment step: in the data processing stage, synthetic aperture radar data is used to assist in determining the water body boundary range; in the dynamic flux accounting stage, the carbon dioxide partial pressure estimation model for the water surface is simplified by setting the chlorophyll a concentration-related terms to zero, so that the carbon dioxide partial pressure estimation model retains only terms related to water surface temperature (i.e., constant terms, first-order terms, and second-order terms).

[0186] In step S300, when calculating the gas transport rate, an ice cap coverage factor is introduced to characterize the inhibitory effect of the ice cap on water-gas interface exchange. The method for obtaining the ice cover area is as follows: based on the differences in backscattering characteristics of the aforementioned synthetic aperture radar (SAR) data, or combined with optical remote sensing imagery, the ice cover area within the target water area is extracted, and the ratio of this ice cover area to the total area of ​​the target water area is calculated. The calculation process for the corrected gas transport rate is expressed by the following formula:

[0187]

[0188] in, This indicates the corrected gas transport rate. This refers to the constant proportionality coefficient determined through calibration. This indicates the wind speed at a height of 10 meters. The Schmitt number, representing the temperature-dependent relationship, is calculated based on the measured water temperature using a conventional empirical polynomial (e.g., ...). ,in The temperature was calculated in Celsius. The value of the ice cover factor is 0 to 1, where 0 indicates no ice cover and 1 indicates complete freezing.

[0189] After completing the adaptive calculations based on different scenarios and outputting the endogenous net carbon sink in step S400, in order to define the cascade errors caused by sensor measurements, atmospheric correction, and model fitting, the following steps S500 and S600 are performed:

[0190] Step S500: Using the Monte Carlo simulation method, apply random perturbations conforming to their typical error distributions to the key input parameters of remote sensing reflectance, water depth, wind speed, and atmospheric carbon dioxide concentration. The calculation process for applying the random perturbations is expressed by the following formula:

[0191]

[0192] in, Indicates the first Perturbation values ​​of key input parameters generated in the Monte Carlo simulation; This represents the actual observed value or the original inversion value of the key input parameter obtained from steps S100 to S300; Indicates the first The random error term generated in this simulation is obtained by random sampling based on the prior uncertainty distribution of each parameter (such as a normal distribution or a log-normal distribution). The relative error coefficient of variation of the prior uncertainty distribution is set according to the following rules: remote sensing reflectance parameter is set to 5%~10%, wind speed parameter to 10%~15%, water depth parameter to 5%, and atmospheric carbon dioxide concentration parameter to 1%~2%**.

[0193] Step S600: By repeatedly executing the complete accounting process, the probability distribution statistical characteristics of endogenous net carbon sinks are obtained, thereby quantifying and outputting the uncertainty of the final assessment result. The calculation process of the uncertainty assessment index is expressed by the following formula:

[0194]

[0195]

[0196]

[0197] in, Indicates the process The expected mean of endogenous net carbon sink obtained from the simulation calculation; This indicates the total number of Monte Carlo simulations set; Net_Intrinsic_Sinkj represents the endogenous net carbon sink output under the j-th perturbation simulation calculation; The standard deviation of the simulation results; This represents the percentage of relative uncertainty in the quantified endogenous net carbon sink at a 95% confidence level. This ensures that the output carbon sink assessment indicators possess rigorous statistical confidence and engineering reference value.

[0198] This ensures that the output carbon sink assessment indicators have rigorous statistical confidence and engineering reference value.

[0199] It should be understood that, in the embodiments of the present invention, "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0200] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0201] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0202] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0203] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0204] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0205] From the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented in hardware, firmware, or a combination thereof. When implemented in software, the above-described functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. For example, but not limited to, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer. Furthermore, any connection can suitably be a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used in this invention, disk and disc include compressed optical discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically magnetically copy data, while discs optically copy data using lasers. The combinations described above should also be included within the scope of protection for computer-readable media.

[0206] In summary, the above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calculating carbon sequestration in rivers and lakes based on remote sensing data, characterized in that, The method includes: S100. Acquire remote sensing images of the target water body covering the visible to thermal infrared bands in the first and second phases within the set evaluation time period, as well as synchronous meteorological data, and perform correction and registration processing to obtain the remote sensing reflectance of the water body and the matching wind speed and atmospheric carbon dioxide concentration data. S200. Based on the remote sensing reflectance of the first and second time phases respectively, the spatial distribution of the concentration of dissolved organic carbon and particulate organic carbon in the water body is retrieved. Combined with the water depth data of the target water area, the vertical integral calculation is performed to obtain the total carbon storage in the water column per unit area under the corresponding time phase. S300. Based on the remote sensing images, the water surface temperature and chlorophyll a concentration are retrieved respectively. The carbon dioxide partial pressure of the water surface is determined using an estimation model. The carbon dioxide partial pressure difference at the water-air interface between the water body and the atmosphere is calculated. The gas transport rate is calculated based on the wind speed. The carbon dioxide exchange flux at the water-air interface is obtained by combining the carbon dioxide solubility constant and the carbon dioxide partial pressure difference at the water-air interface. S400. Accumulate the carbon dioxide exchange flux in time and space within the set evaluation period, and calculate the change in total carbon storage per unit area of ​​water column and the amount of exogenous carbon input within the same time period by subtracting the total carbon storage in the second time phase from the total carbon storage in the first time phase. Obtain the flow rate and river carbon concentration of the inflow section of the target water body to calculate the amount of exogenous carbon input. Establish a collaborative accounting equation for coupled calculation, and output the endogenous net carbon sink generated by the biogeochemical processes inside the water body after deducting the contribution of external input, so as to determine the net carbon source and sink attributes of the water body.

2. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, In step S200, different bio-optical inversion paths are adaptively selected based on a preset water turbidity discrimination threshold. For turbid water, an inversion model based on a combination index is used to invert the dissolved organic carbon concentration, and the inversion process is expressed by the following formula: ； ； Indicates the portfolio index. , and These represent the remote sensing reflectance of the first, second, and third bands, respectively. Indicates the concentration of dissolved organic carbon. Represents the regression coefficient. Indicates the intercept; When retrieving particulate organic carbon concentration, a normalized difference index is constructed and the inversion is performed using an exponential function. The calculation process is expressed by the following formula: ； ； in Indicates the normalized difference index of particulate organic carbon and These represent the remote sensing reflectance of the fourth and fifth bands, respectively. Indicates particulate organic carbon concentration. and This represents the model constant coefficients determined through calibration using measured data.

3. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, In step S200, when obtaining the total carbon storage per unit area of ​​water column, it is assumed that the carbon component concentration decreases exponentially with increasing water depth, and the stratified vertical integral calculation process of the total carbon storage per unit area of ​​water column is expressed by the following formula: ； ； ； in Indicates passing through the sixth band and the seventh band The effective water depth is estimated from the relationship between remote sensing reflectance. and Indicates calibration parameters, Indicates water depth The concentration of carbon components at that location, This represents the concentration of dissolved organic carbon or particulate organic carbon in the surface layer obtained from remote sensing inversion. This represents the diffuse attenuation coefficient estimated from the slope of the remotely sensed reflectance spectrum. This represents the total carbon storage per unit area of ​​water column. This indicates the total number of layers after rounding up the effective water depth. and They represent the first The dissolved organic carbon concentration and particulate organic carbon concentration were calculated using the exponential decay law. Indicates the first The thickness of the layer.

4. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, In step S300, the water surface temperature is retrieved using a split-window algorithm and the chlorophyll a concentration is retrieved using a three-band model. Specifically, this includes: retrieving the water surface temperature using a split-window algorithm based on the brightness temperature of adjacent thermal infrared channels, and expressing the calculation process of the water surface temperature using the following formula: ； The chlorophyll a concentration was retrieved using a three-band model based on the separation of fluorescence signals from red and near-infrared reflectance, and the calculation process for extracting the fluorescence peak height was expressed by the following formula: ； in, This represents the water surface temperature obtained through inversion. and These represent the brightness temperatures obtained after radiometric calibration and atmospheric correction for the first and second thermal infrared channels, respectively. and Indicates the atmospheric water vapor correction factor; Furthermore, the process of obtaining the atmospheric water vapor correction coefficient includes: Using the radiative transfer model, the atmospheric profile parameters during transit are input, and the atmospheric transmittance of the first thermal infrared channel and the second thermal infrared channel is calculated and output. Based on the atmospheric transmittance, interpolation is performed in a pre-established lookup table to obtain the corresponding atmospheric water vapor correction coefficient. and ; This indicates the height of the fluorescence peak located near the ninth band. and These represent the remote sensing reflectance of the eighth, ninth, and tenth bands, respectively.

5. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, In step S300, the estimation model for estimating the partial pressure of carbon dioxide in the water surface is a nonlinear multiple regression model, and its calculation process is expressed by the following formula: ； in, This indicates the partial pressure of carbon dioxide in the surface layer of water. This indicates the inverted water surface temperature. This represents the inverted chlorophyll a concentration. This represents the regression coefficients of the model determined by fitting the training dataset.

6. The method for calculating carbon sequestration in rivers and lakes according to claim 5, characterized in that, In step S300, the calculation process of carbon dioxide exchange flux at the water-air interface is expressed by the following formula: ； in, This represents the carbon dioxide exchange flux. This represents the gas transport rate calculated based on wind speed and Schmidt number. This represents the solubility constant of carbon dioxide in water. This represents the difference in carbon dioxide partial pressure at the water-air interface calculated from the partial pressure of carbon dioxide in water and the partial pressure of carbon dioxide in the atmosphere.

7. The method for calculating carbon sequestration in rivers and lakes according to claim 6, characterized in that, When applied to the calculation of water bodies in high latitudes during the ice-covered period, the method further includes an adaptive adjustment step for the ice-covered period: using synthetic aperture radar data to help determine the boundary range of the water body; The carbon dioxide partial pressure estimation model for the water surface layer is simplified by setting the chlorophyll a concentration-related terms to zero, so that the carbon dioxide partial pressure estimation model retains only the terms related to water surface temperature. In step S300, an ice cover factor is introduced to characterize the inhibitory effect of the ice cover on water-air interface exchange, and the calculation process of the corrected gas transport rate is expressed by the following formula: ； in, This indicates the corrected gas transport rate. The constant proportionality coefficient is determined through calibration. This indicates the wind speed at a height of 10 meters. The Schmidt number represents the relationship between water temperature and temperature. The value of the ice cover factor is 0 to 1, where 0 indicates no ice cover and 1 indicates complete freezing.

8. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, In step S400, the collaborative accounting equation for deducting external input contributions is expressed by the following formula: ； in, This indicates the amount of endogenous net carbon sequestration. This represents the cumulative carbon dioxide exchange flux within the set evaluation period. This indicates the change in the total carbon storage of the water column within the same time period. This indicates the amount of exogenous carbon input from outside the watershed into the target water area, and all parameters are uniformly converted to the same unit of carbon molarity. Furthermore, the calculation process for the amount of exogenous carbon input is expressed by the following formula: ； in, This represents the total amount of exogenous carbon input obtained per unit area of ​​the target water body. Indicates the first The flow rate at the inflow section of the main stream, Indicates the first River carbon concentration at the main stream cross section This represents the total time of the evaluation time window. This indicates the total area of ​​the target water body.

9. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, The method further includes: S500 employs the Monte Carlo simulation method to apply random perturbations to key input parameters such as remote sensing reflectance, water depth, wind speed, and atmospheric carbon dioxide concentration, which conform to their typical error distribution. S600 obtains the probability distribution statistical characteristics of endogenous net carbon sink by repeatedly executing the complete accounting process, thereby quantifying and outputting the uncertainty of the final assessment result.

10. The method for calculating carbon sequestration in rivers and lakes according to claim 1, characterized in that, When applied to the accounting of small bodies of water, the method includes the following adaptive optimization process: Remote sensing data acquired using an unmanned aerial vehicle (UAV) platform was used as the primary data source, and an adapted atmospheric correction model was employed. High-precision water depth data is obtained using airborne lidar technology; A model for estimating the partial pressure of carbon dioxide in water is obtained by training localized samples established for the target water body; and the evaluation time window is shortened accordingly to capture carbon dynamic changes over shorter periods.

Images