Rapid remote sensing identification and flux estimation method and system for near-surface methane abnormal emissions

Abstract

The present invention provides a method and system for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions. The method comprises: calculating the unit absorption characteristic spectra of near-surface methane and other trace gases using a radiative transfer model; selecting the optimal channel for inverting near-surface methane increments; constructing a set of spectrally similar pixels to the pixels to be inverted and a reference background spectrum; performing optimized iterative inversion of near-surface methane increments using a matched filter method; identifying methane point source outlets and methane point source emission plumes using morphological calculations; and calculating the methane point source emission flux based on wind field information at the methane point source outlets. The method can be used for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions, particularly resolving the key technical issue of false alarms in remote sensing identification of abnormal near-surface methane emissions.

Description

Technical Field

[0001] The present invention relates to the field of satellite remote sensing technology, and in particular to a method and system for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions. Background Art

[0002] Methane is a significant greenhouse gas, with a greenhouse effect 25 times greater than that of carbon dioxide per unit concentration. It is also the second most long-lived greenhouse gas after carbon dioxide. Due to abnormal increases in near-surface methane emissions caused by human activities such as coal and natural gas exploration, production and processing, waste disposal, agriculture, and wetlands, monitoring and controlling near-surface methane emissions are crucial for improving energy efficiency and reducing environmental pollution. Traditional methane emission monitoring methods, primarily using on-site monitoring or weather balloons, require significant human, material, and time resources, making it difficult to rapidly acquire information on abnormal methane emissions over a large area.

[0003] In recent years, satellite remote sensing technology has enabled rapid and accurate acquisition of atmospheric gas emission information over a wide range of regions. Quantitative monitoring of atmospheric methane concentrations using multispectral, hyperspectral, and thermal infrared imagery from satellite remote sensing platforms has provided strong support for the development of methane detection research. However, satellite remote sensing data captures all energy reflected from the Earth's surface and the atmosphere. This creates complex and variable background interference for atmospheric trace methane. Furthermore, the presence of many other absorption peaks in the near-infrared band can significantly interfere with the detection of atmospheric methane concentrations. Furthermore, ground objects such as vegetation, water bodies, and buildings, which share similar absorption characteristics with methane, can also interfere with methane signal detection to some extent. Finally, flux estimates of near-surface methane emission sources require integration with meteorological forecast data, and uncertainty in these data can further impact flux estimates.

[0004] In response to the above problems, there is an urgent need for a set of technical solutions that can simultaneously solve the following technical problems:

[0005] (1) How to overcome the false alarm problem in remote sensing identification of abnormal near-surface methane emissions;

[0006] (2) How to achieve rapid identification and flux estimation of abnormal methane emissions from space-based remote sensing.

[0007] Patent document CN114088633B (application number: 202111373589.6) discloses a method for identifying and calculating abnormal methane emissions in coal mining areas through coordinated satellite and ground monitoring, which belongs to the field of remote sensing technology. The invention first identifies the methane emission area based on the satellite's spectral data, and identifies the methane emission facilities in combination with high-resolution satellite imagery. Then, a methane cruise observer is used to conduct cruise observations, establish a coordinate system, and use a Gaussian diffusion model to model the methane diffusion. A genetic algorithm is used to settle the parameters in the Gaussian diffusion model, and finally the exact methane emission intensity is obtained. On the one hand, satellite remote sensing can provide nationwide coverage of methane spatial distribution and long-term series change information from a macro perspective, locate high-value methane emission areas, and detect abnormal methane emission signals in coal mining areas; on the other hand, combined with highly flexible vehicle-mounted mobile monitoring, the methane concentration in the coal mining area can be obtained, and the methane emission flux in the coal mining area can be inverted, providing data support for methane emission reduction in the coal mining area. Summary of the Invention

[0008] In view of the defects in the prior art, the purpose of the present invention is to provide a method and system for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions.

[0009] According to the present invention, a method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions is provided, comprising:

[0010] Step S1: using a radiation transfer model to perform a short-wave infrared radiation brightness value simulation experiment to obtain a unit absorption characteristic spectrum of methane and a unit absorption characteristic spectrum of other trace gases;

[0011] Step S2: selecting the optimal inversion channel for methane increment based on the unit absorption characteristic spectrum of methane and the unit absorption characteristic spectra of other trace gases, and extracting the spectral radiance value based on the optimal inversion channel;

[0012] Step S3: constructing a set of spectrally similar pixels to the pixel to be inverted, and generating a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value;

[0013] Step S4: Calculating the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted;

[0014] Step S5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method; adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted; and repeat steps S3 to S5 until the optimization iteration converges;

[0015] Step S6: generating a methane increment standard grid product based on the methane increment inversion result; identifying methane point source emission outlets and methane point source emission plumes based on the methane increment standard grid product and the wind field information in the study area at the time of satellite transit based on a morphological operation method;

[0016] Step S7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

[0017] Preferably, step S1 adopts: selecting a high spatial resolution satellite remote sensing platform with shortwave infrared hyperspectral observation capability, and extracting parameter configuration information of each band of the satellite sensor; changing the near-surface methane concentration level in the methane prior vertical profile based on the band parameters, using a radiation transfer model to conduct a shortwave infrared radiation brightness value simulation experiment, obtaining the satellite sensor received radiation brightness value under different methane concentration backgrounds, and calculating the unit absorption characteristic spectrum of methane gas in each shortwave infrared band based on different methane background concentrations and corresponding shortwave infrared band radiation brightness data using a least squares fitting method.

[0018] Preferably, step S2 adopts: based on the unit absorption characteristic spectra of methane and other trace gases in each shortwave infrared band, the best channel for near-surface methane increment inversion is screened based on the combined bubble sorting method; the shortwave infrared channel reflectivity data suitable for methane inversion is extracted using the screened best inversion channel, and the shortwave infrared reflectivity data is subjected to radiation calibration calculation to obtain the spectral radiation brightness value.

[0019] Preferably, the step S3 adopts:

[0020] Step S3.1: construct a set of spectrally similar pixels to the pixel to be inverted;

[0021] Step S3.2: Based on the spectral radiance values, perform principal component decomposition on the shortwave infrared spectral radiance in the spectrally similar pixel set, and generate a reference background spectrum of the pixel to be inverted according to the cumulative variance explanation rate of the principal components;

[0022] The spectrally similar pixel set of the pixel to be inverted is composed of pixels whose geographic spatial distance, correlation coefficient, spectral angle and overall deviation from the pixel to be inverted meet preset requirements.

[0023] Preferably, step S4 adopts: taking the pixel to be inverted as the center, extracting the radiation brightness values ​​of the neighboring pixels within a certain spatial window range to form a set, calculating the deviation value between the average spectral radiation brightness value of all pixels in the set and the measured spectral radiation brightness value; calculating the error covariance matrix of the pixel to be inverted based on the deviation value between the average spectral radiation brightness value of all pixels in the set and the measured spectral radiation brightness value;

[0024] The calculation formula of the error covariance matrix is:

[0025]

[0026] Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (-μ) is the anomaly value.

[0027] Preferably, the step S5 adopts:

[0028] Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel;

[0029]

[0030] Where α is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition;

[0031] Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted;

[0032] '

[0033] L i =L i -α i μt

[0034] '

[0035] Among them, L i is the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, α i is the methane increment of the ith pixel, μ is the reference background spectrum, and t is the target characteristic spectrum.

[0036] Preferably, the step S6 adopts:

[0037] The methane increment standard grid product is generated by: geometrically correcting the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product;

[0038] The morphological operation method adopts: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots;

[0039] The methane point source emission outlet adopts: identifying the high value center of the patch;

[0040] The methane point source emission plume is determined by calculating the Euclidean distance between each pixel in the image patch and the point source emission port, determining the pixel with the farthest spatial distance, and setting the image patch whose azimuth and wind direction angle between the point source emission port pixel and the farthest pixel is less than a preset value as a methane point source emission plume.

[0041] Preferably, the step S7 adopts:

[0042] The equivalent length of the methane point source emission plume is:

[0043]

[0044] Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel;

[0045] The wind field information of the methane point source emission outlet is:

[0046] U eff =1.1*U+0.5

[0047] Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet;

[0048] The calculation of methane point source emission flux adopts:

[0049] The formula for calculating the integrated mass increment IME of the methane point source emission plume is:

[0050]

[0051] Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i Indicates the methane increment value inverted by the pixel;

[0052] The methane emission flux Q of the methane point source emission plume is calculated using the formula:

[0053]

[0054] Among them, k is the methane unit conversion parameter, A is the geographical coverage area of ​​the pixel, α i is the initial value of methane increment in the i-th pixel; Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, U eff is the equivalent wind speed.

[0055] According to the present invention, a rapid remote sensing identification and flux estimation system for abnormal near-surface methane emissions is provided, comprising:

[0056] Module M1: Use the radiation transfer model to simulate the short-wave infrared radiation brightness value to obtain the unit absorption characteristic spectrum of methane and other trace gases;

[0057] Module M2: Screen the optimal inversion channel for methane increment based on the unit absorption characteristic spectrum of methane and other trace gases, and extract the spectral radiance value based on the optimal inversion channel;

[0058] Module M3: Construct a set of spectrally similar pixels to the pixel to be inverted, and generate a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value;

[0059] Module M4: Calculate the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted;

[0060] Module M5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method. Adjust the spectral radiation brightness value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted. Repeat the triggering of modules M3 to M5 until the optimization iteration converges.

[0061] Module M6: Generate methane increment standard grid products based on methane increment inversion results; identify methane point source emission outlets and methane point source emission plumes based on methane increment standard grid products and wind field information in the study area at the time of satellite transit using morphological operation methods;

[0062] Module M7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

[0063] Preferably, the module M1 adopts: selecting a high spatial resolution satellite remote sensing platform with shortwave infrared hyperspectral observation capability, and extracting parameter configuration information of each band of the satellite sensor; changing the near-surface methane concentration level in the methane prior vertical profile based on the band parameters, using a radiation transfer model to conduct a shortwave infrared radiation brightness value simulation experiment, obtaining the satellite sensor received radiation brightness value under different methane concentration backgrounds, and calculating the unit absorption characteristic spectrum of methane gas in each shortwave infrared band using a least squares fitting method based on different methane background concentrations and corresponding shortwave infrared band radiation brightness data;

[0064] The module M2 adopts: based on the unit absorption characteristic spectra of methane and other trace gases in each short-wave infrared band, the optimal channel for near-surface methane increment inversion is selected based on the combined bubble sorting method; the short-wave infrared channel reflectivity data suitable for methane inversion is extracted using the selected optimal inversion channel, and the short-wave infrared reflectivity data is subjected to radiometric calibration calculation to obtain the spectral radiation brightness value;

[0065] The module M3 adopts:

[0066] Module M3.1: Construct a set of spectrally similar pixels to the pixel to be inverted;

[0067] Module M3.2: Perform principal component decomposition of the shortwave infrared spectral radiance in the spectrally similar pixel set based on the spectral radiance value, and generate a reference background spectrum for the pixel to be inverted based on the cumulative variance explained by the principal components;

[0068] The spectral similar pixel set of the pixel to be inverted is composed of pixels whose geographic space distance, correlation coefficient, spectral angle and overall deviation meet preset requirements with the pixel to be inverted;

[0069] The module M4 adopts the following methods: taking the pixel to be inverted as the center, extracting the radiance values ​​of neighboring pixels within a certain spatial window range to form a set, calculating the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; and calculating the error covariance matrix of the pixel to be inverted based on the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value.

[0070] The calculation formula of the error covariance matrix is:

[0071]

[0072] Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (-μ) is the anomaly value;

[0073] The module M5 adopts:

[0074] Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel;

[0075]

[0076] Where α is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition;

[0077] Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted;

[0078] '

[0079] L i =L i -α i μt

[0080] '

[0081] Among them, L iis the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, α i is the methane increment of the i-th pixel, μ is the reference background spectrum, and t is the target characteristic spectrum;

[0082] The module M6 adopts:

[0083] The methane increment standard grid product is generated by: geometrically correcting the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product;

[0084] The morphological operation method adopts: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots;

[0085] The methane point source emission outlet adopts: identifying the high value center of the patch;

[0086] The methane point source emission plume is determined by calculating the Euclidean distance between each pixel in the image patch and the point source emission outlet, determining the pixel with the farthest spatial distance, and setting the image patch where the angle between the azimuth and wind direction of the point source emission outlet pixel and the farthest pixel is less than a preset value as the methane point source emission plume;

[0087] The module M7 adopts:

[0088] The equivalent length of the methane point source emission plume is:

[0089]

[0090] Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel;

[0091] The wind field information of the methane point source emission outlet is:

[0092] U eff =1.1*U+0.5

[0093] Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet;

[0094] The calculation of methane point source emission flux adopts:

[0095] The formula for calculating the integrated mass increment IME of the methane point source emission plume is:

[0096]

[0097] Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i Indicates the methane increment value inverted by the pixel;

[0098] The methane emission flux Q of the methane point source emission plume is calculated using the formula:

[0099]

[0100] Among them, k is the methane unit conversion parameter, A is the geographical coverage area of ​​the pixel, α i is the initial value of methane increment in the i-th pixel; Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, U eff is the equivalent wind speed.

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] 1. The present invention solves the key technical problem of reconstructing the near-surface methane unit absorption characteristic spectrum and reference background spectrum, reduces the false alarm rate of remote sensing identification of abnormal near-surface methane emissions, and realizes the rapid identification and flux estimation of abnormal methane emissions by space-based remote sensing;

[0103] 2. The present invention proposes an adaptive reference pixel screening and reference spectrum construction method to reduce the false alarm problem in remote sensing identification of abnormal near-surface methane emissions;

[0104] 3. The present invention proposes a method for rapid identification of abnormal methane plumes based on morphological processing, which realizes the automated identification and flux estimation of abnormal methane emissions from space-based remote sensing. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0106] Figure 1 This is a flow chart of a method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions.

[0107] Figure 2 It is a schematic diagram of the result of methane anomaly increment inversion using the present invention.

[0108] Figure 3 It is a schematic diagram of the result of estimating abnormal methane emission flux using the present invention. DETAILED DESCRIPTION

[0109] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0110] Example 1

[0111] According to the present invention, a method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions is provided. Figure 1-3 Shown, including:

[0112] Step S1: using a radiation transfer model to perform a short-wave infrared radiation brightness value simulation experiment to obtain a unit absorption characteristic spectrum of methane and a unit absorption characteristic spectrum of other trace gases; wherein the radiation transfer model has a spectrum line-by-line integration characteristic;

[0113] Step S2: selecting the optimal inversion channel for methane increment based on the unit absorption characteristic spectrum of methane and the unit absorption characteristic spectra of other trace gases, and extracting the spectral radiance value based on the optimal inversion channel;

[0114] Step S3: constructing a set of spectrally similar pixels to the pixel to be inverted, and generating a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value;

[0115] Step S4: Calculating the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted;

[0116] Step S5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method; adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted; and repeat steps S3 to S5 until the optimization iteration converges;

[0117] Step S6: generating a methane increment standard grid product based on the methane increment inversion result; identifying methane point source emission outlets and methane point source emission plumes based on the methane increment standard grid product and the wind field information in the study area at the time of satellite transit based on a morphological operation method;

[0118] Step S7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

[0119] Specifically, step S1 adopts the following methods: selecting a high spatial resolution satellite remote sensing platform with shortwave infrared hyperspectral observation capability, and extracting parameter configuration information of each band of the satellite sensor; changing the near-surface methane concentration level in the methane prior vertical profile based on the band parameters, and using a radiation transfer model to simulate the shortwave infrared radiation brightness value to obtain the satellite sensor received radiation brightness value under different methane concentration backgrounds; and using the least squares fitting method to calculate the unit absorption characteristic spectrum of methane gas in each shortwave infrared band based on different methane background concentrations and corresponding shortwave infrared band radiation brightness data.

[0120] Specifically, step S2 adopts: based on the unit absorption characteristic spectra of methane and other trace gases in each shortwave infrared band, the optimal channel for near-surface methane increment inversion is screened based on the combined bubble sorting method; the shortwave infrared channel reflectivity data suitable for methane inversion is extracted using the screened optimal inversion channel, and the shortwave infrared reflectivity data is subjected to radiation calibration calculation to obtain the spectral radiation brightness value.

[0121] Specifically, the step S3 adopts:

[0122] Step S3.1: construct a set of spectrally similar pixels to the pixel to be inverted;

[0123] Step S3.2: Based on the spectral radiance values, perform principal component decomposition on the shortwave infrared spectral radiance in the spectrally similar pixel set, and generate a reference background spectrum of the pixel to be inverted according to the cumulative variance explanation rate of the principal components;

[0124] The spectrally similar pixel set of the pixel to be inverted is composed of pixels whose geographic spatial distance, correlation coefficient, spectral angle and overall deviation from the pixel to be inverted meet preset requirements.

[0125] Specifically, step S4 adopts: taking the pixel to be inverted as the center, extracting the radiance values ​​of neighboring pixels within a certain spatial window range to form a set, calculating the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; calculating the error covariance matrix of the pixel to be inverted based on the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value;

[0126] The calculation formula of the error covariance matrix is:

[0127]

[0128] Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (-μ) is the anomaly value.

[0129] Specifically, the step S5 adopts:

[0130] Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel;

[0131]

[0132] Where α is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition;

[0133] Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted;

[0134] '

[0135] L i =L i -α i μt

[0136] '

[0137] Among them, L i is the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, α i is the methane increment of the ith pixel, μ is the reference background spectrum, and t is the target characteristic spectrum.

[0138] Specifically, the step S6 adopts:

[0139] The methane increment standard grid product is generated by: geometrically correcting the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product;

[0140] The morphological operation method adopts: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots;

[0141] The methane point source emission outlet adopts: identifying the high value center of the patch;

[0142] The methane point source emission plume is determined by calculating the Euclidean distance between each pixel in the image patch and the point source emission port, determining the pixel with the farthest spatial distance, and setting the image patch whose azimuth and wind direction angle between the point source emission port pixel and the farthest pixel is less than a preset value as a methane point source emission plume.

[0143] Specifically, the step S7 adopts:

[0144] The equivalent length of the methane point source emission plume is:

[0145]

[0146] Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel;

[0147] The wind field information of the methane point source emission outlet is:

[0148] U eff =1.1*U+0.5

[0149] Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet;

[0150] The calculation of methane point source emission flux adopts:

[0151] The formula for calculating the integrated mass increment IME of the methane point source emission plume is:

[0152]

[0153] Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i Indicates the methane increment value inverted by the pixel;

[0154] The methane emission flux Q of the methane point source emission plume is calculated using the formula:

[0155]

[0156] Among them, k is the methane unit conversion parameter, A is the geographical coverage area of ​​the pixel, α i is the initial value of methane increment in the i-th pixel; Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, U eff is the equivalent wind speed.

[0157] According to the present invention, a rapid remote sensing identification and flux estimation system for abnormal near-surface methane emissions is provided, comprising:

[0158] Module M1: Using a radiation transfer model to simulate the brightness of short-wave infrared radiation, obtain the unit absorption characteristic spectrum of methane and other trace gases; wherein the radiation transfer model has a spectrum line-by-line integration characteristic;

[0159] Module M2: Screen the optimal inversion channel for methane increment based on the unit absorption characteristic spectrum of methane and other trace gases, and extract the spectral radiance value based on the optimal inversion channel;

[0160] Module M3: Construct a set of spectrally similar pixels to the pixel to be inverted, and generate a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value;

[0161] Module M4: Calculate the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted;

[0162] Module M5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method. Adjust the spectral radiation brightness value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted. Repeat the triggering of modules M3 to M5 until the optimization iteration converges.

[0163] Module M6: Generate methane increment standard grid products based on methane increment inversion results; identify methane point source emission outlets and methane point source emission plumes based on methane increment standard grid products and wind field information in the study area at the time of satellite transit using morphological operation methods;

[0164] Module M7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

[0165] Specifically, the module M1 adopts: selecting a high-spatial-resolution satellite remote sensing platform with short-wave infrared hyperspectral observation capabilities, and extracting parameter configuration information of each band of the satellite sensor; changing the near-surface methane concentration level in the methane prior vertical profile based on the band parameters, and using the radiation transfer model to simulate the short-wave infrared radiation brightness value to obtain the satellite sensor received radiation brightness value under different methane concentration backgrounds; according to different methane background concentrations and corresponding short-wave infrared band radiation brightness data, using the least squares fitting method, calculate the unit absorption characteristic spectrum of methane gas in each short-wave infrared band.

[0166] Specifically, the module M2 adopts: according to the unit absorption characteristic spectra of methane and other trace gases in each short-wave infrared band, based on the combined bubble sorting method, the optimal channel for near-surface methane increment inversion is screened; the short-wave infrared channel reflectivity data suitable for methane inversion is extracted using the screened optimal inversion channel, and the short-wave infrared reflectivity data is calculated through radiation calibration to obtain the spectral radiation brightness value.

[0167] Specifically, the module M3 adopts:

[0168] Module M3.1: Construct a set of spectrally similar pixels to the pixel to be inverted;

[0169] Module M3.2: Perform principal component decomposition of the shortwave infrared spectral radiance in the spectrally similar pixel set based on the spectral radiance value, and generate a reference background spectrum for the pixel to be inverted based on the cumulative variance explained by the principal components;

[0170] The spectrally similar pixel set of the pixel to be inverted is composed of pixels whose geographic spatial distance, correlation coefficient, spectral angle and overall deviation from the pixel to be inverted meet preset requirements.

[0171] Specifically, the module M4 adopts: taking the pixel to be inverted as the center, extracting the radiance values ​​of the neighboring pixels within a certain spatial window range to form a set, calculating the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; calculating the error covariance matrix of the pixel to be inverted based on the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value;

[0172] The calculation formula of the error covariance matrix is:

[0173]

[0174] Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (-μ) is the anomaly value.

[0175] Specifically, the module M5 adopts:

[0176] Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel;

[0177]

[0178] Where α is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition;

[0179] Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted;

[0180] '

[0181] L i =L i -α i μt

[0182] '

[0183] Among them, L i is the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, α i is the methane increment of the ith pixel, μ is the reference background spectrum, and t is the target characteristic spectrum.

[0184] Specifically, the module M6 adopts:

[0185] The methane increment standard grid product is generated by: geometrically correcting the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product;

[0186] The morphological operation method adopts: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots;

[0187] The methane point source emission outlet adopts: identifying the high value center of the patch;

[0188] The methane point source emission plume is determined by calculating the Euclidean distance between each pixel in the image patch and the point source emission port, determining the pixel with the farthest spatial distance, and setting the image patch whose azimuth and wind direction angle between the point source emission port pixel and the farthest pixel is less than a preset value as a methane point source emission plume.

[0189] Specifically, the module M7 adopts:

[0190] The equivalent length of the methane point source emission plume is:

[0191]

[0192] Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel;

[0193] The wind field information of the methane point source emission outlet is:

[0194] U eff =1.1*U+0.5

[0195] Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet;

[0196] The calculation of methane point source emission flux adopts:

[0197] The formula for calculating the integrated mass increment IME of the methane point source emission plume is:

[0198]

[0199] Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i Indicates the methane increment value inverted by the pixel;

[0200] The methane emission flux Q of the methane point source emission plume is calculated using the formula:

[0201]

[0202] Among them, k is the methane unit conversion parameter, A is the geographical coverage area of ​​the pixel, α i is the initial value of methane increment in the i-th pixel; Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, U eff is the equivalent wind speed.

[0203] Example 2

[0204] Example 2 is a preferred example of Example 1

[0205] According to the present invention, a method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions is provided, comprising:

[0206] Step 1: Using the radiation transfer model, a method for simulating the near-surface methane unit absorption characteristic spectrum is proposed to screen the optimal channel for inversion of atmospheric methane anomaly increment and perform matched filter inversion.

[0207] Specifically, the step 1 includes:

[0208] Step 1.1: Select a hyperspectral satellite remote sensing platform (with a spectral range covering the 1.67 μm and 2.3 μm bands with typical methane absorption characteristics) and obtain the satellite sensor's band configuration information, including the center wavelength, full-width at half maximum, and spectral response function of each band.

[0209] Step 1.2: Set the simulation parameters of the radiative transfer model (including but not limited to the SCIATRAN and LBLTRAN models), including: satellite sensor spectral characteristics (central wavelength, full width at half maximum, etc.), satellite observation conditions (geometric parameters of the sun and satellite), atmospheric parameters (aerosol parameterization scheme, aerosol optical depth, vertical profile of absorbing gases), surface reflectivity, and surface elevation.

[0210] Step 1.3: Change the near-surface methane concentration in the methane prior vertical profile and simulate the radiance values ​​observed by satellite sensors under different concentration backgrounds;

[0211] Step 1.4: Based on the simulation results, the absorption rate of methane in each short-wave infrared band is calculated by the least squares fitting method based on the input of the prior methane background concentration and the corresponding simulated radiation brightness data of each band, and the unit absorption characteristic spectrum of methane s is obtained. 1×b .

[0212] Step 2: Simulate the unit absorption characteristic spectra of other trace gases separately and use the combined bubble sorting method to screen the optimal channel for near-surface methane increment inversion.

[0213] Step 2.1: Using the same simulation method as Step 1, calculate the unit absorption characteristic spectra of other absorbing gases (water vapor, nitrogen oxides, nitrous oxide, nitrogen dioxide, and carbon dioxide) in the same shortwave infrared band. Specifically, set the simulation parameters of the radiative transfer model, including: satellite sensor spectral characteristics (central wavelength, full-width at half maximum), satellite observation conditions (solar-satellite geometry), atmospheric parameters (aerosol parameterization scheme, aerosol optical depth, vertical profile of absorbing gases), surface reflectivity, and surface elevation. Vary the prior vertical profile concentrations of other trace gases and simulate the radiance values ​​observed by the satellite sensor under different concentration backgrounds. Trace gases include water vapor, nitrogen oxides, nitrous oxide, nitrogen dioxide, and carbon dioxide.

[0214] Step 2.2: Combine the calculated methane unit absorption characteristic spectrum and use the combined bubble sorting method to screen out the bands with strong methane absorption characteristics and weak absorption characteristics of other gases;

[0215] Step 2.3: Based on the results of the combined bubble sort method, determine the optimal channel for near-surface methane increment inversion within the shortwave infrared band range. The number of bands is marked as b.

[0216] Step 3: Construct a set of spectrally similar pixels to the pixel to be inverted and generate a reference background spectrum for the pixel to be inverted.

[0217] Step 3.1: Extract the shortwave infrared L1 reflectance data of the hyperspectral sensor and perform radiometric calibration on each band of data according to the corresponding calibration coefficient to generate a radiometric brightness value with physical meaning;

[0218] Step 3.2: Use a sliding spectrum window to smooth the spectral radiance value of each pixel after calibration;

[0219] Step 3.3: For each pixel to be inverted, calculate the spectral angle, correlation coefficient, overall deviation, and spatial distance between its spectral vector and the spectral vectors of other pixels in the same scene image;

[0220] Step 3.4: Use the non-dominated sorting method to select the first 100 pixels with close geographic distance to the pixel to be inverted, high correlation coefficient, small spectral angle and overall deviation to form a spectrally similar pixel set;

[0221] Step 3.5: Perform principal component decomposition on the set of spectrally similar pixels and calculate the overall variance explanation rate of each principal component;

[0222] Step 3.6: Based on the spatial sparseness assumption of point source emissions, select the first n principal components with a cumulative variance explanation rate exceeding 90% for matrix reconstruction to generate the reference background spectrum μ of the pixel to be inverted 1×b .

[0223] Step 4: Calculate the error covariance matrix of the pixel to be inverted using the spectral radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted.

[0224] Step 4.1: With the pixel to be inverted as the center, extract the spectral radiance values ​​of the corresponding pixels within the 25×25 spatial window to form a set X 25×25×b ;

[0225] Step 4.2: Calculate the set X 25×25×b Average spectrum, denoted as μ′ 1×b ;

[0226] Step 4.3: Calculate the set X 25×25×b Measured spectral radiance L 1×b and the average spectrum μ′ 1×b The anomaly value of

[0227] Step 4.4: Calculate the error covariance matrix C using the anomaly spectrum values b×b .

[0228] Step 5: Use the matched filter method to perform iterative optimization inversion of near-surface methane increment.

[0229] Step 5.1: For the pixel to be inverted, calculate its observed radiance L 1×b Compared with the reference background spectrum μ 1×b The difference between

[0230] L′ 1×b =L 1×b -μ 1×b

[0231] Step 5.2: Using the reference background spectrum μ 1×b Unit absorption characteristic spectrum of methane 1×b Stretch, calculate the dot product of the two, and update the target characteristic spectrum t′ 1×b ;

[0232] t′ 1×b =s 1×b ⊙μ 1×b

[0233] Step 5.3: Use the matched filter method to invert the initial value of methane increment pixel by pixel;

[0234]

[0235] Step 5.4: Adjust the spectral radiance value of each pixel in the similar pixel set according to the initial value α of the methane increment; repeat steps 3 to 5 to update the reference background spectrum μ and the error covariance matrix C; perform iterative optimization of the methane increment inversion until the iteration converges;

[0236] L′ i =L i -α i μt

[0237] Step 6: Based on the morphological operation method and the methane increment inversion results, identify the methane point source emission plume.

[0238] Step 6.1: Use the geographic location information of each pixel in the original observation data to perform geometric correction on the methane increment inversion results to generate a methane increment standard grid product;

[0239] Step 6.2: Obtain wind field information (i.e., wind speed and direction data) within the study area at the time of satellite transit;

[0240] Step 6.3: Perform Gaussian smoothing filtering on the methane increment standard grid product;

[0241] Step 6.4: Use the Sobel algorithm to extract the boundary of the methane increment standard grid product after Gaussian smoothing filtering;

[0242] Step 6.5: Perform morphological dilation on the boundary extraction results to fill the gaps in the image.

[0243] Step 6.6: Erode the result of step 6.5 in a diamond pattern to remove isolated pixels.

[0244] Step 6.7: Perform image closing operation on the result of step 6.6 in an eight-neighborhood manner to generate independent image patches;

[0245] Step 6.8: Remove small patches with less than 50 pixels.

[0246] Step 6.9: Identify the high-value center of the patch and set it as the methane point source emission outlet;

[0247] Step 6.10: Calculate the Euclidean distance between each pixel in the image patch and the methane point source emission outlet, and determine the pixel with the farthest spatial distance;

[0248] Step 6.11: Calculate the azimuth angle between the point source outlet pixel and the pixel with the farthest spatial distance;

[0249] Step 6.12: Calculate the angle between the azimuth and the wind direction;

[0250] Step 6.13: Define the image patches with angles less than 90° as methane point source emission plumes;

[0251] Step 7: Combined with meteorological wind field information, calculate the methane point source emission flux.

[0252] Step 7.1: For each identified methane point source emission plume, calculate the geographic area of ​​its pixel and record it as A.

[0253] Step 7.2: Calculate the equivalent length L of the methane point source emission plume;

[0254]

[0255] Step 7.3: Calculate the equivalent wind speed U based on the meridional wind speed U at the methane point source outlet. eff ;

[0256] U eff =1.1*U+0.5

[0257] Step 7.4: Calculate the integrated mass increment IME of each methane point source emission plume, where k is the methane unit conversion parameter;

[0258]

[0259] Step 7.5: Calculate the methane emission flux Q of the methane point source emission plume;

[0260]

[0261] Those skilled in the art will appreciate that, in addition to implementing the system, device, and various modules provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like by logically programming the method steps. Therefore, the system, device, and various modules provided by the present invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; the modules for implementing various functions can also be considered both software programs for implementing the method and structures within the hardware component.

[0262] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions, characterized in that: include: Step S1: using a radiation transfer model to perform a short-wave infrared radiation brightness value simulation experiment to obtain the unit absorption characteristic spectra of methane and the unit absorption characteristic spectra of other trace gases in each short-wave infrared band; Step S2: screening the optimal inversion channel for methane increment using a combined bubble sorting method based on the unit absorption characteristic spectrum of methane and the unit absorption characteristic spectra of other trace gases, and extracting the spectral radiance value based on the optimal inversion channel; Step S3: constructing a set of spectrally similar pixels to the pixel to be inverted, and generating a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value; the set of spectrally similar pixels to the pixel to be inverted is composed of pixels whose geographic spatial distance, correlation coefficient, spectral angle, and overall deviation meet preset requirements with the pixel to be inverted; Step S4: Calculating the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted; Step S5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method; adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted; and repeat steps S3 to S5 until the optimization iteration converges; Step S6: generating a methane increment standard grid product based on the methane increment inversion result; identifying methane point source emission outlets and methane point source emission plumes based on the methane increment standard grid product and the wind field information in the study area at the time of satellite transit based on a morphological operation method; Step S7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

2. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 1 is characterized in that: The step S1 comprises: selecting a high spatial resolution satellite remote sensing platform with shortwave infrared hyperspectral observation capability, and extracting parameter configuration information of each band of the satellite sensor; Based on the band parameters, the near-surface methane concentration level in the methane prior vertical profile is changed, and a shortwave infrared radiation brightness value simulation experiment is carried out using the radiation transfer model to obtain the radiation brightness value received by the satellite sensor under different methane concentration backgrounds. According to the different methane background concentrations and the corresponding shortwave infrared band radiation brightness data, the least squares fitting method is used to calculate the unit absorption characteristic spectrum of methane gas in each shortwave infrared band.

3. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 1 is characterized in that: The step S2 adopts: using the selected best inversion channel to extract short-wave infrared channel reflectance data suitable for methane inversion, and the short-wave infrared reflectance data is subjected to radiation calibration calculation to obtain the spectral radiation brightness value.

4. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 1, characterized in that: The step S3 adopts: Step S3.1: construct a set of spectrally similar pixels to the pixel to be inverted; Step S3.2: Based on the spectral radiance values, perform principal component decomposition on the shortwave infrared spectral radiance in the spectrally similar pixel set, and generate a reference background spectrum of the pixel to be inverted according to the cumulative variance explanation rate of the principal component.

5. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 1, characterized in that: The step S4 adopts the following steps: taking the pixel to be inverted as the center, extracting the radiance values ​​of the neighboring pixels within a certain spatial window range to form a set, calculating the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; calculating the error covariance matrix of the pixel to be inverted based on the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; The calculation formula of the error covariance matrix is: Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (L-μ) is the anomaly value.

6. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 5, characterized in that: The step S5 adopts: Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel; Where a is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition; Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted; L i ′=L i -a i ·μ·t Among them, L i ′ is the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, a i is the methane increment of the ith pixel, μ is the reference background spectrum, and t is the target characteristic spectrum.

7. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 1, characterized in that: The step S6 adopts: Generating the methane increment standard grid product includes: performing geometric correction on the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product; The morphological operation method includes: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots; Identify the high-value center of the patch as a methane point source emission outlet; The Euclidean distance between each pixel in the patch and the point source emission outlet is calculated to determine the pixel with the farthest spatial distance. The patch whose azimuth angle and wind direction angle between the point source emission outlet pixel and the farthest pixel is less than a preset value is set as a methane point source emission plume.

8. The method for rapid remote sensing identification and flux estimation of abnormal near-surface methane emissions according to claim 7, characterized in that: The step S7 adopts: The equivalent length of the methane point source emission plume is: Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel; The wind field information of the methane point source emission outlet is: IN eff =1.1*U+0.5 Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet; The calculation of methane point source emission flux adopts: The formula for calculating the integrated mass increment IME of the methane point source emission plume is: Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i represents the methane increment value inverted by the i-th pixel; The methane emission flux Q of the methane point source emission plume is calculated using the formula: Where Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, and U is eff is the equivalent wind speed.

9. A rapid remote sensing identification and flux estimation system for abnormal near-surface methane emissions, characterized by: include: Module M1: Use the radiation transfer model to simulate the shortwave infrared radiation brightness value to obtain the unit absorption characteristic spectra of methane and other trace gases in each shortwave infrared band; Module M2: using a combined bubble sorting method to select the optimal inversion channel for methane increment based on the unit absorption characteristic spectrum of methane and the unit absorption characteristic spectra of other trace gases, and extracting the spectral radiance value based on the optimal inversion channel; Module M3: Constructing a set of spectrally similar pixels to the pixel to be inverted, and generating a reference background spectrum of the pixel to be inverted based on the constructed set of spectrally similar pixels and the spectral radiance value; the set of spectrally similar pixels to the pixel to be inverted is composed of pixels whose geographic spatial distance, correlation coefficient, spectral angle, and overall deviation meet preset requirements with the pixel to be inverted; Module M4: Calculate the error covariance matrix of the pixel to be inverted based on the radiance values ​​of other pixels within a certain spatial window of the pixel to be inverted; Module M5: Calculate the methane increment of the pixel to be inverted based on the reference background spectrum of the pixel to be inverted and the error covariance matrix of the pixel to be inverted using the matched filter method. Adjust the spectral radiation brightness value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted. Repeat the triggering of modules M3 to M5 until the optimization iteration converges. Module M6: Generate methane increment standard grid products based on methane increment inversion results; identify methane point source emission outlets and methane point source emission plumes based on morphological operation methods using methane increment standard grid products and wind field information in the study area at the time of satellite transit; Module M7: Calculate the methane point source emission flux based on the methane point source emission plume and the wind field information of the methane point source emission outlet.

10. The rapid remote sensing identification and flux estimation system for abnormal near-surface methane emissions according to claim 9, characterized in that: The module M1 adopts: selecting a high spatial resolution satellite remote sensing platform with short-wave infrared hyperspectral observation capability and extracting parameter configuration information of each band of the satellite sensor; Based on the band parameters, the near-surface methane concentration level in the methane prior vertical profile was changed. A shortwave infrared radiation brightness value simulation experiment was conducted using a radiation transfer model to obtain the satellite sensor's received radiation brightness value under different methane concentration backgrounds. Based on the different methane background concentrations and the corresponding shortwave infrared band radiation brightness data, the least squares fitting method was used to calculate the unit absorption characteristic spectrum of methane gas in each shortwave infrared band. The module M2 adopts: using the selected best inversion channel to extract the short-wave infrared channel reflectance data suitable for methane inversion, and the short-wave infrared reflectance data is subjected to radiation calibration calculation to obtain the spectral radiation brightness value; The module M3 adopts: Module M3.1: Construct a set of spectrally similar pixels to the pixel to be inverted; Module M3.2: Perform principal component decomposition of the shortwave infrared spectral radiance in the spectrally similar pixel set based on the spectral radiance value, and generate a reference background spectrum for the pixel to be inverted based on the cumulative variance explained by the principal components; The module M4 adopts the following methods: taking the pixel to be inverted as the center, extracting the radiance values ​​of neighboring pixels within a certain spatial window range to form a set, calculating the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value; and calculating the error covariance matrix of the pixel to be inverted based on the deviation between the average spectral radiance value of all pixels in the set and the measured spectral radiance value. The calculation formula of the error covariance matrix is: Where N is the number of pixels in the set, L is the measured spectral radiance value, μ is the average spectral radiance value, and (L-μ) is the anomaly value; The module M5 adopts: Based on the matched filter method, the methane increment of the pixel to be inverted is calculated through the reference spectrum of each pixel; Where a is the initial value of the methane increment; L′ is the difference vector between the observed radiance and the pixel reference background spectrum; C is the error covariance matrix; t′ is the target characteristic spectrum obtained by stretching and updating the methane unit absorption characteristic spectrum using the reference background spectrum; T represents transposition; Adjust the spectral radiance value of each pixel in the spectrally similar pixel set based on the methane increment of the pixel to be inverted; L i ′=L i -a i ·μ·t Among them, L i ′ is the adjusted spectral radiance value of the i-th pixel, L i is the observed radiance of the i-th pixel, a i is the methane increment of the i-th pixel, μ is the reference background spectrum, and t is the target characteristic spectrum; The module M6 adopts: Generating the methane increment standard grid product includes: performing geometric correction on the methane increment inversion result to generate the methane increment standard grid product, and performing Gaussian smoothing filtering on the methane increment standard grid product; The morphological operation method includes: using the Sobel algorithm to extract the boundary, then performing dilation processing in a morphological manner, performing erosion processing in a diamond manner, and performing image closing operation processing in an eight-neighborhood manner to achieve the purpose of removing small spots; Identify the high-value center of the patch as a methane point source emission outlet; Calculating the Euclidean distance between each pixel in the patch and the point source emission outlet, determining the pixel with the farthest spatial distance, and setting the patch where the angle between the azimuth and wind direction of the point source emission outlet pixel and the farthest pixel is less than a preset value as a methane point source emission plume; The module M7 adopts: The equivalent length of the methane point source emission plume is: Where L is the equivalent length of the methane point source emission plume, and A is the geographical coverage area of ​​the pixel; The wind field information of the methane point source emission outlet is: IN eff =1.1*U+0.5 Among them, U eff is the equivalent wind speed, U is the meridional wind speed at the methane point source emission outlet; The calculation of methane point source emission flux adopts: The formula for calculating the integrated mass increment IME of the methane point source emission plume is: Among them, k represents the methane unit conversion parameter; N represents the number of pixels; α i represents the methane increment value inverted by the i-th pixel; The methane emission flux Q of the methane point source emission plume is calculated using the formula: Where Q is the methane emission flux of the methane point source emission plume, L is the equivalent length of the plume, and U is eff is the equivalent wind speed.

Images