Method for estimating dissolved methane concentration of algae type lake water body based on satellite remote sensing image

Abstract

The invention relates to a satellite remote sensing image-based method for estimating the concentration of dissolved methane in an algae-type lake water body. The concentration of dissolved CH4 in the water body is estimated through a lake water environment parameter satellite remote sensing product. A satellite remote sensing image is used for processing and analyzing the dissolved CH4 concentration of the lake water body every to two days, the spatial distribution of the dissolved CH4 concentration of the lake water body is obtained based on pixel calculation, the dissolved CH4 concentration of the water body of the whole lake and sub-regions of the lake is subjected to statistical analysis, and remote sensing automatic estimation of the dissolved CH4 of the lake water body is preliminarily realized. By adopting the method disclosed by the invention, the spatial distribution of the dissolved CH4 concentration of the lake water body with high spatial resolution can be comprehensively obtained, the method has important reference and guiding significance for evaluating the CH4 emission of the lake water-gas interface, and important scientific basis and technical support can be provided for regional carbon emission and environmental protection.

Description

Technical Field

[0001] The present invention relates to the technical fields of environmental engineering and image processing, and particularly relates to a method for estimating the dissolved methane concentration in the water body of an algal lake based on satellite remote sensing images. Background Art

[0002] Lakes are important indicators of climate change, providing a variety of information-rich signals about the physical, chemical, and biological responses to climate. Lakes are also key regulatory hubs for the terrestrial carbon cycle and the core nodes for the conversion of carbon fluxes at the atmosphere-hydrosphere-land interface. Despite their small size, these aquatic systems can affect the regional carbon balance, and their annual methane (CH4) emissions account for approximately half of the global anthropogenic and natural methane emissions. More alarmingly, the eutrophication process caused by human activities is rapidly amplifying this ecological effect. Model estimates suggest that the increased CH4 release may result in global socio-economic costs of up to $8.1 trillion. However, due to the internal heterogeneity of aquatic ecosystems and the spatial heterogeneity among river basins, there are still significant uncertainties in the assessment of CH4 emissions from inland waters such as lakes in the current carbon accounting system, which has become a key scientific problem restricting the accurate measurement of the global carbon budget.

[0003] Currently, the assessment of lake CH4 emissions faces significant methodological bottlenecks: the existing monitoring systems generally have the dual limitations of incomplete observation time series and insufficient sampling resolution, resulting in significant biases in the estimation of emission fluxes. The traditional research paradigm relies on discretized field sampling at large spatial scales, and the limited spatial sampling density is difficult to capture the spatial distribution of CH4 in the whole lake. The low-frequency snapshot sampling also cannot resolve the seasonal dynamic characteristics of emission fluxes. In theory, to achieve a relatively accurate assessment of lake CH4 emissions, at least seasonal field sampling is required, and sampling sites need to be evenly distributed across the whole lake. This resource-intensive method is restricted by the discreteness of the distribution of lake water bodies and is difficult to implement in the assessment of lake CH4 emissions at the regional or global scale.

[0004] Satellite remote sensing has the advantages of large scale and periodicity, which can make up for the deficiencies of traditional field surveys of lake CH4 concentration in terms of spatial scale and long-term assessment. At present, satellite remote sensing is mainly used to estimate the CH4 concentration in the atmosphere, while there are more difficulties in estimating the dissolved methane concentration in water bodies. The concentration of methane dissolved in water is extremely low, and coupled with the strong absorption characteristics of the water body itself, the difficulty of estimating the dissolved methane concentration in water bodies is significantly increased. The applicant previously tried to use satellite remote sensing products to estimate the methane flux at the water-air interface in the Taihu Lake area at the micromole level. Nevertheless, the dissolved methane concentration in water bodies has a smaller magnitude compared with the methane flux, about the nanomole level, and is more difficult to accurately estimate. At the same time, the interaction of factors controlling the methane production and consumption processes in algal lakes may be more complex between lakes and within lake controls, making it difficult to achieve unified remote sensing estimation of water body methane concentration in large areas of multiple lakes. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for estimating the dissolved CH4 concentration in the water body of algal lakes based on satellite remote sensing images, which uses existing publicly available satellite remote sensing products of lake water environment parameters to accurately estimate the dissolved CH4 concentration and its spatial variation in the target water area, providing important scientific basis and technical support for regional carbon emissions and environmental protection.

[0006] To achieve the above purpose, the technical solutions adopted by the present invention are as follows:

[0007] A method for estimating the dissolved methane concentration in the water body of algal lakes based on satellite remote sensing images, the method comprising:

[0008] Obtaining the average water depth, chlorophyll a concentration Chla, water temperature, and water transparency Z of the target water area based on satellite remote sensing data SD and instantaneous photosynthetically active radiation iPAR data;

[0009] Estimating the daily cumulative photosynthetically active radiation AccuPAR of the target water area based on the iPAR data;

[0010] Using the average water depth, Chla, water temperature, Z SD and AccuPAR as independent variables and the measured methane concentration of the target water area as the dependent variable to construct a regression equation;

[0011] Estimating the dissolved methane concentration in the water body of algal lakes based on the regression equation.

[0012] In some embodiments of the present invention, the water temperature and iPAR data are obtained based on remote sensing product data, and the Z SD and chlorophyll a concentration are retrieved based on remote sensing image data.

[0013] In some embodiments of the present invention, the satellite remote sensing data is processed by removing bloom pixels and aquatic vegetation pixels. Preferably, the FAI index is used to remove bloom pixels and aquatic vegetation pixels.

[0014] In some embodiments of the present invention, the estimation method of AccuPAR is as follows:

[0015] Based on the geographical location and date of the target water area, the sunrise and sunset times of the target water area are calculated;

[0016] Combined with the iPAR product data and the satellite observation time, the diurnal variation curve of iPAR from sunrise to sunset is fitted by a quadratic function, and the AccuPAR is obtained by integrating the daily variation curve of iPAR.

[0017] In some embodiments of the present invention, the dependent variable and the independent variable are paired data that have been spatio-temporally matched; the spatio-temporal matching includes:

[0018] The time difference between the sampling time of the sampling point and the satellite observation time does not exceed a preset time;

[0019] At least half of the pixels in the n×n window around the central pixel of the sampling point coordinates have valid data, and the coefficient of variation of these valid data is less than a preset value.

[0020] In some embodiments of the present invention, the method further includes arranging the paired data in the order of methane concentration, and selecting one sample every m samples as a verification sample to construct a training set and a verification set.

[0021] In some embodiments of the present invention, the independent variable is processed by standardization; and the Chla, Z SD data is processed by logarithmization, and a regression equation with ln(Chla), water temperature, ln(Z SD ), AccuPAR, average depth and as the dependent variable is established.

[0022] In some embodiments of the present invention, the form of the regression equation is a multivariate quadratic equation.

[0023] Further, the established regression equation is as follows:

[0024]

[0025] In the formula, x1 = ln(Chla), x2 = water temperature, x3 = ln(Z SD ), x4 = average water depth, x5 = AccuPAR, A0 - A 12 are the equation coefficients.

[0026] In some embodiments of the present invention, the measured methane concentration in the target water area is the methane concentration of the surface water body sample in the target water area. Preferably, the methane concentration is measured based on the headspace equilibrium method.

[0027] In some embodiments of the present invention, the water temperature and iPAR data are obtained based on the Sentinel 3A / 3B satellite standard L2-level remote sensing product data;

[0028] The Z SD and chlorophyll a concentration are retrieved from the OLCI-Sentinel 3A / 3B remote sensing image data.

[0029] Furthermore, the method further includes calculating the dissolved CH4 concentration in the entire lake water body to generate a dissolved CH4 concentration image of the water body, which serves as data support for long-term temporal and spatial variation analysis.

[0030] The algorithm principle of the present invention is as follows:

[0031] Theoretically, although most of the methane in lakes comes from anaerobic production in bottom sediments, in eutrophic algal lakes, a large part of the organic matrix required for methane production in sediments comes from surface algae. At the same time, the methane production process is significantly affected by temperature, and the environmental conditions on the surface and bottom of shallow lakes are relatively similar. Therefore, the temperature at the bottom of the lake can be estimated using the temperature of the surface water body. In addition, oxygen is produced through photosynthesis by phytoplankton in surface water, which promotes the oxidation of methane and thus affects the methane concentration in the surface water body. The intensity of photosynthesis is controlled by the chlorophyll a concentration, photosynthetically active radiation, and water transparency in the water body. Although the estimated methane concentration is somewhat underestimated, it is basically consistent with the measured results in terms of spatial distribution and numerical value. The vertical exchange of lake water bodies easily leads to the resuspension of sediments, which not only transports methane at the bottom to the surface water body but also causes a concomitant change in water transparency. Therefore, water transparency can also indirectly reflect the relative level of methane in the surface water body. The transport process of methane from the bottom to the surface water body is directly controlled by the lake depth. A longer transport process makes it more likely for methane to be oxidized and also increases the process cost of transporting relevant substances and energy to the sediments. In summary, the present invention is based on satellite remote sensing Chla, Tw, Z SD and PAR products, supplemented by the average water depth as an auxiliary variable, and a quantitative relationship between these independent variables and the dissolved CH4 concentration in the water body is established through a multiple quadratic polynomial regression algorithm, realizing the remote sensing estimation of the dissolved CH4 concentration in the water body.

[0032] The method of the present invention can quickly obtain the spatial distribution of dissolved CH4 concentration in lake water bodies with high precision and high spatial resolution based on remote sensing data, which has important reference and guiding significance for evaluating CH4 emissions at the lake water-air interface, and can provide important scientific basis and technical support for regional carbon emissions and environmental protection.

[0033] It should be understood that all combinations of the foregoing concepts and additional concepts described in more detail below can be regarded as part of the inventive subject matter of the present disclosure as long as such concepts do not conflict with each other. In addition, all combinations of the claimed subject matter are regarded as part of the inventive subject matter of the present disclosure.

[0034] The foregoing and other aspects, embodiments and features of the teachings of the present invention can be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the present invention, such as the features and / or beneficial effects of exemplary embodiments, will be apparent from the following description, or will be learned through the practice of specific embodiments in accordance with the teachings of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in each figure. Now, embodiments of various aspects of the present invention will be described by way of example and with reference to the drawings, wherein:

[0036] Figure 1 is a scatter plot of the model-estimated and measured CH4 concentrations in Example 1, where the dots represent the training set samples and the diamonds represent the validation set samples.

[0037] Figure 2 is the true color image of the Hongze Lake OLCI image on April 17, 2021 (left) and the estimated result of the CH4 concentration distribution (right) in Example 1.

[0038] Figure 3 is the true color image of the Honghu Lake OLCI image on December 21, 2021 (left) and the estimated result of the CH4 concentration distribution (right) in Example 1.

[0039] Figure 4 is the true color image of the Chaohu Lake OLCI image on December 18, 2021 (left) and the estimated result of the CH4 concentration distribution (right) in Example 1.

[0040] Figure 5 is a schematic diagram of the daily average and monthly average changes in the dissolved CH4 concentration in the whole lakes of Hongze Lake, Honghu Lake and Chaohu Lake from May 2016 to December 2021 in Example 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] To better understand the technical content of the present invention, specific embodiments are hereby given and described in conjunction with the accompanying drawings as follows.

[0042] In the present disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which numerous illustrative embodiments are shown. The embodiments of the present disclosure are not necessarily intended to cover all aspects of the present invention. It should be understood that the various concepts and embodiments introduced above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed in the present invention are not limited to any particular implementation. Additionally, some aspects of the present invention can be used alone or in any suitable combination with any other aspects of the present invention.

[0043] Example 1

[0044] In this example, a typical algae-type lake in the Yangtze-Huaihe River Basin is taken as an example, and the technical solution of the present invention is further described using Sentinel-3A / B satellite data.

[0045] The data sources in the example are as follows:

[0046] According to the size, hydrological conditions, and nutrient status of the lakes, 16 typical algae-type lakes were selected in the middle and lower reaches of the Yangtze-Huaihe River Basin, and six field samplings were carried out in July 2018, October 2019, September and October 2020, and April, June, and December 2021. In each sampling, 5 to 30 sampling points were set in the open water area of each lake and synchronized with the Sentinel-3 satellite overpass. At each sampling point, surface water samples with a depth of 0 to 50 cm were collected to obtain the dissolved CH4 concentration and supporting water quality parameters. The headspace equilibrium method was used to measure the methane concentration in the water body. Approximately 16 ml of bubble-free surface water sample was injected into a 30 ml glass serum bottle through a 20 ml polypropylene syringe. In the laboratory, a gas chromatograph (Agilent-7890B) was used to measure the volume fraction of CH4 gas in the headspace gas. According to the measured volume fraction of the headspace gas, the average total pressure at the sampling point, the in-situ water temperature, the laboratory equilibrium temperature, and the Henry's law constant, the CH4 concentration under in-situ conditions was calculated.

[0047] Download the OLCI-Sentinel-3 A / B L1B FR data covering the middle and lower reaches of the Yangtze River Basin and the land surface temperature of the Sea and Land Surface Temperature Radiometer (SLSTR) L2 as input products for the lake surface water temperature Tw from the European Space Agency (ESA) Copernicus Data Space System (https: / / dataspace.copernicus.eu / ). The original resolution of the land surface temperature product is 1 km, and it is resampled to a resolution of 300 m to match other OLCI sensor data products. Download the OLCI-Sentinel-3 A / B L2 instantaneous photosynthetically active radiation (iPAR) product from the European Organization for the Exploitation of Meteorological Satellites (EUMETSAT, https: / / data.eumetsat.int / ). Download the MYD04 aerosol optical depth (AOD) daytime product with a resolution of 3 km of the MODIS satellite for the corresponding area of each lake from the NASA EOSDIS (https: / / earthdata.nasa.gov) website, and calculate the average aerosol optical depth of the corresponding area of the lake. Based on the 6SV atmospheric correction algorithm and combined with the obtained aerosol optical depth, perform atmospheric correction on the OLCI-Sentinel-3 L1B data to obtain the OLCI remote sensing reflectance (Rrs) dataset. Using the OLCI Rrs data, invert the lake surface water Chla product based on the machine learning algorithm proposed by Shen et al. (2020) (Shen, M., Luo, J., Cao, Z., Xue, K., Qi, T., Ma, J., Liu, D., Song, K., Feng, L., & Duan, H. (2022). Random forest: an optimal chlorophyll-a algorithm for optically complex inland water suffering atmospheric correction uncertainties. Journal of Hydrology , 128685); Invert the lake water body Z based on the machine learning algorithm proposed by Shen et al. (2020) SDProduct (Shen, M., Duan, H., Cao, Z., Xue, K., Qi, T., Ma, J., Liu, D., Song, K., Huang, C., & Song, X. (2020). Sentinel-3 OLCI observations of water clarity in large lakes in eastern China: Implications for SDG 6.3.2 evaluation. Remote Sensing of Environment, 247 )

[0048] Through the HydroLAKES lake hydrological dataset (https: / / www.hydrosheds.org / products / hydrolakes), its "Depth_avg" data was extracted to obtain the average water depth Depth of the target water area. A corresponding rasterized layer was generated based on the spatial resolution of OLCI-Sentinel-3 A / B, and the obtained average water depth Depth was assigned to all pixels within the water area at the same time.

[0049] After that, the estimation of dissolved CH4 concentration in the Taihu Lake water body was based on satellite Chla, Tw, Z SD , and iPAR product data. The specific processing flow is as follows:

[0050] 1) Preprocess the remotely sensed Chla, Tw, Z SD , and iPAR product data. Since meaningful Chla and Z SD parameters cannot be retrieved in the cyanobacterial bloom area, the floating algae index (FAI) was used to remove the bloom pixels in the image data (FAI > 1.194×10 -4 ). In addition, since there is a large amount of aquatic vegetation in the eastern region of the Taihu Lake, the method described in the present invention is not applicable, and the spectral characteristics of aquatic vegetation are similar to those of algal blooms. Therefore, this part of the data was also removed by masking with the FAI index.

[0051] 2) Since the diurnal variation of iPAR follows a quadratic function (R² = 0.65, p < 0.01), we used the quadratic function to fit the diurnal variation of iPAR based on the satellite observation time, Sentinel-3 iPAR product, sunrise, and sunset times. The sunrise and sunset times were calculated from the date, longitude, and latitude of each lake. The daily cumulative PAR (AccuPAR) was obtained by integrating the daily iPAR curve from sunrise to sunset;

[0052] 3) Standardize all input variables, and for Chla, Z SDThe product data is logarithmically processed with the natural constant e as the base;

[0053] 4) Using the criteria that ① the time difference between the sampling time and the satellite observation time does not exceed 5 hours, and ② at least half of the pixels in the 3×3 pixel window around the central pixel of the sampling point coordinates have valid data, and the coefficient of variation of these valid data is less than 10%, match the water body CH4 concentration of the field real sample points with the pixel data of the preprocessed remote sensing image raster data to obtain a sample-pixel paired dataset. A total of 246 sample-pixel pairs were obtained. Arrange the paired dataset in ascending order of CH4 concentration, and select one sample every 3 samples as the validation sample to obtain a model training dataset with 149 paired data and a model validation dataset with 97 paired data.

[0054] 5) Perform regression calculations through the samples of the model training dataset to determine the equation coefficients. The regression equation is as follows:

[0055]

[0056] In the formula, x1 = ln(Chla), x2 = Tw, x3 = ln(Z SD ), x4 = Depth, x5 = AccuPAR, A0 - A 12 are the equation coefficients, A0 = 5.56, A1 = 0.47, A2 = -0.98, A3 = 0.16, A4 = 1.43, A5 = -0.21, A6 = -1.69, A7 = -0.45, A8 = -0.52, A9 = 1.32, A 10 = 0.13, A 11 = -0.26, A 12 = 0.78. This equation is established based on the sampling data of multiple typical lakes in the middle and lower reaches of the Yangtze River in the embodiment and is also applicable to similar algae-type lakes in this basin. For lake groups in other basins, it is necessary to re-calibrate the coefficients An of each term based on the lake data.

[0057] Evaluate the estimation accuracy of the model using the samples of the training dataset and the validation dataset. The results are as Figure 1 shown. The root mean square error (RMSE) and mean absolute percentage error (MAPE) between the estimated results of the model and the field measured CH4 concentration results are within an acceptable range, and the scatter points are evenly distributed on both sides of the 1:1 line, indicating that the model has good estimation ability for CH4 concentrations in the range of 10 - 1000 nmol L -1 close to 2 orders of magnitude.

[0058] 6) Associate the preprocessed remote sensing image raster data through the longitude and latitude grid, and use pixels as the unit for Chla, Tw, Z SD, substitute the Depth and AccuPAR data into the regression equation to estimate the dissolved CH4 concentration (cCH4, nmol L -1 ) of the corresponding pixel.

[0059] 7) Calculate statistical indicators such as the spatial distribution of the monthly average CH4 concentration over the years and the average value and standard deviation of the dissolved CH4 concentration in the entire lake water body for typical lakes, as data support for the long-term time-series spatio-temporal variation analysis.

[0060] Finally, post-processing of the dissolved CH4 concentration image data of the lake water body: including generating the dissolved CH4 concentration image of the water body and obtaining its spatial distribution.

[0061] Using the above algorithm, automatically calculate the dissolved CH4 concentration of the lake water body in the middle and lower reaches of the Yangtze River and Huaihe River basins, and find that the algorithm plays a good role for images of different lake waters and different water body characteristics. The estimation results of Hongze Lake, Honghu Lake and Chaohu Lake are specifically as Figures 2 - 4 shown. Figure 5 Further statistical analysis of the daily average and monthly average CH4 concentration change curves of Hongze Lake, Honghu Lake and Chaohu Lake over a long time series is carried out to analyze the spatio-temporal differentiation law of dissolved CH4 in the water body.

[0062] Through the above method, an estimation model of the dissolved CH4 in the lake water body based on remote sensing images can be established, and the estimation accuracy is relatively high; accordingly, satellite remote sensing monitoring will have good prospects in the research on the carbon cycle and greenhouse gas emissions of large algal lakes. The present invention can comprehensively obtain the spatial distribution of the dissolved CH4 concentration in the lake water body with a spatial resolution of 300 m, which has important reference and guiding significance for evaluating the CH4 emission at the water-air interface of the lake, and can provide important scientific basis and technical support for regional carbon emissions and environmental protection.

[0063] Example 2

[0064] In this example, multiple regression models are constructed based on different input parameter combinations for methane concentration estimation, and the accuracies of multiple regression models are compared. The Akaike Information Criterion (AIC) is used to compare the goodness of fit between each model; the coefficient of determination (R 2 ) is used to compare the degree of explanation of each model for the change of methane concentration; the estimation performance of each model is compared through the estimation accuracies of the training set and the validation set samples, such as the root mean square error (RMSE), the absolute value of the average relative error (MAPE), and the average ratio (MR).

[0065] The input parameters are mainly the potential influencing factors of the methane concentration in the lake, such as chlorophyll a concentration (Chla), water temperature (Tw), water transparency (Z SD), average water depth (Depth), daily cumulative photosynthetically active radiation (AccuPAR), time corresponding to the observation time (Ts), daily cumulative photosynthetically active radiation amount at the observation time (aPAR), and the proportion of the cumulative radiation amount at the observation time in the total daily radiation amount (ks). Among them, based on the satellite observation time, Sentinel-3 iPAR product, sunrise and sunset times, the diurnal variation of iPAR following a quadratic function shape was fitted. Sunrise and sunset times were calculated from the date, longitude, and latitude of each lake. AccuPAR was obtained by integrating the daily iPAR curve from sunrise to sunset; aPAR could be obtained by integrating the iPAR curve from sunrise to the observation time; ks was the ratio of aPAR to AccuPAR.

[0066] The results are shown in Table 1. The input parameter combinations include the combination considering only the parameters of the lake water body itself (A1), and the estimation models obtained from the combinations A2 to A5 formed by gradually adding external radiation factors of the lake. The performance of the A1 model in both the training set and the validation set was not ideal. The A2 model added the AccuPAR input parameter on the basis of the A1 model, and its performance in both the training set and the validation set was significantly improved. When more complex models were formed by continuing to add input parameters (A3 - A5), there was no obvious improvement in the goodness of fit and explanatory degree of the model itself for the methane model, while the estimation error showed a significant increase, indicating that the model had overfitting phenomenon and poor generalization ability. Therefore, the combination of input parameters of the methane estimation model proposed in this method has a robust estimation and explanatory ability for the dynamic change of methane concentration in algal lakes, while maintaining a relatively simple model form, and retains good generalization potential for extended applications in lakes in other regions.

[0067] Table 1 Precision index table of methane concentration estimation models with different input parameter combinations

[0068]

[0069] Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those with ordinary knowledge in the technical field to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to what is defined by the claims.

Claims

1. A method for estimating the dissolved methane concentration in the water body of algal lakes based on satellite remote sensing images, characterized in that, The method includes: Obtain the average water depth, chlorophyll a concentration Chla, water temperature, and water transparency Z of the target water area based on satellite remote sensing data SD , and instantaneous photosynthetically active radiation iPAR data; estimating the daily cumulative photosynthetically active radiation AccuPAR of the target water area based on the iPAR data; Using the average water depth, Chla, water temperature, Z SD and AccuPAR as independent variables, and the measured methane concentration in the target water area as the dependent variable, a regression equation is constructed; estimating the dissolved methane concentration in the water body of the algal lake based on the regression equation.

2. The method according to claim 1, characterized in that, The water temperature and iPAR data are obtained based on remote sensing product data, and the SD chlorophyll a concentration is retrieved based on remote sensing image data.

3. The method according to claim 1, characterized in that, The satellite remote sensing data is processed by removing the water bloom pixels and aquatic vegetation pixels.

4. The method according to claim 1, characterized in that, The estimation method of the AccuPAR is as follows: calculating the sunrise and sunset times of the target water area based on the geographical location and date of the target water area; combining the iPAR product data and the satellite observation time, fitting the daily variation curve of iPAR from sunrise to sunset using a quadratic function, and obtaining the AccuPAR by integrating the daily variation curve of iPAR.

5. The method according to claim 1, characterized in that, The dependent variable and the independent variable are paired data that have undergone spatio-temporal matching; the spatio-temporal matching includes: the time difference between the sampling time of the sampling point and the satellite observation time does not exceed a preset time; Around the central pixel of the sampling point coordinates n × n At least half of the pixels in the window have valid data, and the coefficient of variation of these valid data is less than a preset value.

6. The method according to claim 5, wherein It also includes arranging the paired data in the order of methane concentration magnitude, and picking out one sample every m samples as a validation sample to construct a training set and a validation set.

7. The method according to claim 1, wherein The independent variable is subjected to standardization processing; and the Chla and Z SD data are subjected to logarithmic processing.

8. The method according to claim 1, wherein the form of the regression equation is a multivariate quadratic equation.

9. The method according to claim 1, characterized in that The measured methane concentration in the target water area is the methane concentration of the surface water body sample in the target water area.

10. The method according to claim 1 or 2, characterized in that The water temperature and iPAR data are obtained based on the Sentinel3A / 3B satellite standard L2-level remote sensing product data; The said Z SD and chlorophyll a concentration are retrieved based on OLCI-Sentinel 3A / 3B remote sensing image data.

Images