Remote sensing monitoring method and device for sulfur dioxide in atmosphere, equipment and medium

Abstract

The invention discloses a remote sensing monitoring method and device for sulfur dioxide in atmosphere, equipment and a medium, and relates to the technical field of remote sensing detection. The method comprises the following steps: indexing from a target data set obtained by performing space-time analysis processing on an initial data set to obtain a high-resolution satellite remote sensing spectrum, space-time dimension data after data preprocessing and wavelength dimension data, performing space dimension splitting and pixel marking on spectrum data in a CCD row and column range corresponding to a to-be-inverted region according to the space-time dimension data, and obtaining a high-resolution satellite remote sensing spectrum; performing parallel processing on the obtained multiple partition data blocks; performing spectrum dimension splitting on the observation spectrum of each partition data block based on the wavelength dimension data, and inputting a plurality of obtained sub-waveband data blocks into a radiation transmission model in parallel to obtain a simulation spectrum corresponding to each sub-waveband data block; and performing fitting and data integration on each simulation spectrum and observation spectrum, and correcting the obtained initial inversion result of the sulfur dioxide concentration to obtain a target inversion result. And the sulfur dioxide inversion calculation efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of remote sensing detection technology, and in particular to a remote sensing monitoring method, device, equipment and medium for sulfur dioxide in the atmosphere. Background Art

[0002] In the field of atmospheric environment monitoring, remote sensing technology is becoming an indispensable monitoring method. Remote sensing technology provides information about the earth's surface and atmosphere by detecting and analyzing the electromagnetic wave information reflected or radiated by the earth's surface and atmosphere. In particular, satellite remote sensing technology based on passive observation is widely used in atmospheric environment monitoring due to its wide coverage, rapid information acquisition capabilities, and the characteristics of not being restricted by terrain areas. Among them, atmospheric sulfur dioxide (SO2) is one of the six typical polluting gases, so monitoring SO2 is a key part of environmental monitoring. The early remote sensing payloads had unsatisfactory temporal and spatial resolution and accuracy due to the limitations of instrument performance and observation technology. With the development of technology, the resolution and signal-to-noise ratio of the most advanced satellite-borne sensors have been significantly improved, providing unprecedented atmospheric monitoring capabilities, but at the same time, higher requirements have been placed on data processing capabilities.

[0003] At present, the inversion algorithm of atmospheric sulfur dioxide mainly relies on the Lambert-Beer law combined with the radiation transfer model for calculation, including nonlinear fitting methods. Although the nonlinear fitting algorithm can provide more accurate inversion results, it requires multiple spectral simulations, which leads to low computational efficiency, especially when facing large-scale data, its processing speed is difficult to meet the needs of business operations. Therefore, how to improve the efficiency of sulfur dioxide inversion calculation is an urgent problem to be solved. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a remote sensing monitoring method, device, equipment, and medium for sulfur dioxide in the atmosphere, which can invert the sulfur dioxide concentration in the atmosphere through a segmentation strategy and a parallel strategy based on the spatiotemporal dimension and the spectral wavelength dimension, thereby improving the sulfur dioxide inversion calculation efficiency.

[0005] In a first aspect, an embodiment of the present application provides a remote sensing monitoring method for sulfur dioxide in the atmosphere, comprising:

[0006] Perform spatiotemporal analysis on the initial data set obtained from monitoring to obtain the target data set;

[0007] According to the time parameter, high-resolution satellite remote sensing spectrum, spatiotemporal dimension data and wavelength dimension data after data preprocessing are obtained from the target data set by indexing;

[0008] Determine the CCD row and column range corresponding to the area to be inverted according to the latitude and longitude information of the high-resolution satellite remote sensing spectrum;

[0009] According to the time-space dimension data after data preprocessing, the spectral data of the CCD row and column range corresponding to the inversion area is subjected to space dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions;

[0010] Performing parallel data processing and spectral preprocessing on the multiple partitioned data blocks, performing spectral dimension splitting processing on the observed spectrum of each of the partitioned data blocks based on the wavelength dimension data, to obtain multiple sub-band data blocks;

[0011] Inputting the plurality of sub-band data blocks into the radiation transfer model in parallel, performing simulation parallel calculation processing based on the simulation spectrum parameters, and obtaining a simulation spectrum corresponding to each of the sub-band data blocks;

[0012] Performing fitting processing and data integration processing on each of the simulated spectra and the observed spectra to obtain an initial inversion result of sulfur dioxide concentration;

[0013] The initial inversion result is subjected to stripe background correction processing and dimensional background correction processing in sequence to obtain a target inversion result of the corrected sulfur dioxide concentration.

[0014] In a second aspect, an embodiment of the present application provides a remote sensing monitoring device for sulfur dioxide in the atmosphere, comprising at least one processor and a memory for communicating with the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the remote sensing monitoring method for sulfur dioxide in the atmosphere as described in any one of the embodiments of the first aspect.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, comprising a remote sensing monitoring device for sulfur dioxide in the atmosphere as described in the embodiment of the second aspect.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute a remote sensing monitoring method for sulfur dioxide in the atmosphere as described in any one of the embodiments of the first aspect.

[0017] The embodiment of the present application includes: by using a remote sensing monitoring device for sulfur dioxide in the atmosphere, after performing spatiotemporal analysis processing on the initial data set obtained by monitoring to obtain a target data set, first, according to the time parameter, a high-resolution satellite remote sensing spectrum, spatiotemporal dimension data after data preprocessing, and wavelength dimension data are indexed from the target data set; secondly, according to the longitude and latitude information of the high-resolution satellite remote sensing spectrum, the CCD row and column range corresponding to the area to be inverted is determined; then, according to the spatiotemporal dimension data after data preprocessing, the spectral data of the CCD row and column range corresponding to the area to be inverted is subjected to spatial dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions; then, the multiple partition data blocks are parallelly processed. According to data processing and spectral preprocessing, the observed spectrum of each partitioned data block is subjected to spectral dimension splitting processing based on the wavelength dimension data to obtain multiple sub-band data blocks; then, the multiple sub-band data blocks are input into the radiation transfer model in parallel, and simulated parallel calculation processing is performed based on the simulated spectral parameters to obtain the simulated spectrum corresponding to each sub-band data block; then, each simulated spectrum and the observed spectrum are subjected to fitting processing and data integration processing to obtain the initial inversion result of sulfur dioxide concentration; finally, the initial inversion result is subjected to stripe background correction processing and dimensional background correction processing in turn to obtain the target inversion result of the corrected sulfur dioxide concentration; by inverting the sulfur dioxide concentration in the atmosphere in parallel, the sulfur dioxide inversion calculation efficiency is improved. That is to say, the embodiment of the present application inverts the sulfur dioxide concentration in the atmosphere through a segmentation strategy and a parallel strategy based on the spatiotemporal dimension and the spectral wavelength dimension, thereby improving the sulfur dioxide inversion calculation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the steps of remote sensing monitoring of sulfur dioxide in the atmosphere provided by an embodiment of the present application;

[0019] Figure 2 yes Figure 1 Specific flow diagram of step S180;

[0020] Figure 3 This is a schematic diagram of the hardware structure of a remote sensing monitoring device for sulfur dioxide in the atmosphere provided by an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0022] It should be noted that although a logical order is shown in the flowchart in the description of the present application, in some cases, the steps shown or described may be performed in an order different from that in the flowchart. In the description of the present application, a number of means one or more, and a plurality of means two or more. The description of "first" and "second" is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0024] First, some terms used in this application are explained:

[0025] Tikhonov regularization, also known as Tikhonov regularization or ridge regression, is a regularization technique widely used in statistics, machine learning, and solving ill-posed problems. It improves the generalization ability of the model by adding a regularization term to the objective function and helps stabilize coefficient estimation, especially on data sets with highly correlated features.

[0026] The Gauss-Newton method is an iterative algorithm for solving nonlinear least squares problems; it is mainly used for parameter estimation problems where the model is a nonlinear function of the parameters and there are differences between the observed data and the model predictions. The Gauss-Newton method minimizes the sum of squared residuals between the observed data and the model predictions by iterative approximation.

[0027] The present application discloses a remote sensing monitoring method and device for sulfur dioxide in the atmosphere, an electronic device and a computer-readable storage medium, and relates to the field of remote sensing detection technology. The method comprises: from the target data set obtained by performing spatiotemporal analysis on the initial data set, indexing and obtaining high-resolution satellite remote sensing spectrum, spatiotemporal dimension data and wavelength dimension data after data preprocessing, performing spatial dimension splitting and pixel marking on the spectrum data of the CCD row and column range corresponding to the inversion area according to the spatiotemporal dimension data, and processing the obtained multiple partition data blocks in parallel; performing spectral dimension splitting on the observed spectrum of each partition data block based on the wavelength dimension data, and inputting the obtained multiple sub-band data blocks into the radiation transmission model in parallel to obtain the simulated spectrum corresponding to each sub-band data block; after fitting and data integration of each simulated spectrum and observed spectrum, correcting the obtained initial inversion result of sulfur dioxide concentration to obtain the target inversion result. It can improve the calculation efficiency of sulfur dioxide inversion.

[0028] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0029] First, as Figure 1 As shown, the remote sensing monitoring method for sulfur dioxide in the atmosphere may include but is not limited to steps S110 to S180.

[0030] Step S110: performing spatiotemporal analysis on the initial data set obtained through monitoring to obtain a target data set.

[0031] Step S120: obtaining high-resolution satellite remote sensing spectrum, spatiotemporal dimension data after data preprocessing, and wavelength dimension data from the target data set according to the time parameter.

[0032] Step S130: Determine the CCD row and column range corresponding to the area to be inverted according to the latitude and longitude information of the high-resolution satellite remote sensing spectrum.

[0033] Step S140: According to the pre-processed spatiotemporal data, the spectral data of the CCD row and column range corresponding to the inversion area is subjected to spatial dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions.

[0034] Step S150: performing parallel data processing and spectral preprocessing on the multiple partitioned data blocks, performing spectral dimension splitting processing on the observed spectrum of each partitioned data block based on the wavelength dimension data, and obtaining multiple sub-band data blocks.

[0035] Step S160: inputting a plurality of sub-band data blocks into the radiation transfer model in parallel, performing simulation parallel calculation processing based on the simulation spectrum parameters, and obtaining a simulation spectrum corresponding to each sub-band data block.

[0036] Step S170: performing fitting processing and data integration processing on each simulated spectrum and the observed spectrum to obtain an initial inversion result of sulfur dioxide concentration.

[0037] Step S180: The initial inversion result is subjected to stripe background correction and dimensional background correction in sequence to obtain a target inversion result of the corrected sulfur dioxide concentration.

[0038] The time parameter in step S120 refers to a selected month or several months, thereby selecting a target data set within the time parameter, for example, selecting a target data set for October, or a target data set within half a year.

[0039] Through steps S110 to S180, by using the remote sensing monitoring device for sulfur dioxide in the atmosphere, after performing spatiotemporal analysis processing on the initial data set obtained by monitoring to obtain the target data set, first, the high-resolution satellite remote sensing spectrum, the spatiotemporal dimension data after data preprocessing, and the wavelength dimension data are indexed from the target data set according to the time parameter; secondly, according to the latitude and longitude information of the high-resolution satellite remote sensing spectrum, the CCD row and column range corresponding to the area to be inverted is determined; then, according to the spatiotemporal dimension data after data preprocessing, the spectral data of the CCD row and column range corresponding to the area to be inverted is subjected to spatial dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions; then, the multiple partition data blocks are subjected to Parallel data processing and spectral preprocessing, based on the wavelength dimension data, the observed spectrum of each partitioned data block is subjected to spectral dimension splitting processing to obtain multiple sub-band data blocks; then, the multiple sub-band data blocks are input into the radiation transfer model in parallel, and simulated parallel calculation processing is performed based on the simulated spectrum parameters to obtain the simulated spectrum corresponding to each sub-band data block; then, each simulated spectrum and the observed spectrum are subjected to fitting processing and data integration processing to obtain the initial inversion result of sulfur dioxide concentration; finally, the initial inversion result is subjected to stripe background correction processing and dimensional background correction processing in turn to obtain the target inversion result of the corrected sulfur dioxide concentration; by inverting the sulfur dioxide concentration in the atmosphere in parallel, the sulfur dioxide inversion calculation efficiency is improved. That is to say, the embodiment of the present application inverts the sulfur dioxide concentration in the atmosphere through a segmentation strategy and a parallel strategy based on the spatiotemporal dimension and the spectral wavelength dimension, thereby improving the sulfur dioxide inversion calculation efficiency.

[0040] It should be noted that the spectral data collected by the satellite has a three-dimensional structure, also known as a data cube, including two spatial dimensions and a single spectral dimension. In view of the characteristics of satellite spectral data, especially the characteristics in the spatial dimension and the spectral dimension, the traditional serial processing method seems to be incapable of doing so, which makes it limited in practical application. The traditional serial calculation method refers to the execution of tasks in a certain order, and each task must wait for the previous task to be completed before it can start. This method is linear, and only one processing unit executes the task. When processing large-scale spectral simulations, a large amount of computing resources are often required, and the calculation process takes a long time. In contrast, the embodiment of the present application significantly improves the processing speed and computing power by using parallel computing and using multiple processors to execute multiple tasks or different parts of a task at the same time. And adopting the principle based on the optimization estimation algorithm, combined with the spatial dimension and spectral dimension characteristics of satellite spectral data, a parallel computing strategy is developed; it can improve the processing speed and the time of simulating the spectrum from the algorithm level, thereby providing support for business data analysis.

[0041] According to some embodiments of the present application, step S110 is further described. Specifically, the initial data set includes: an initial hyperspectral data set and initial inversion auxiliary data; the target data set includes: a hyperspectral data set divided by time series, spatiotemporal dimension data divided by time series, and wavelength dimension data divided by band intervals; step S110: performing spatiotemporal analysis processing on the initial data set obtained by monitoring to obtain a target data set, including but not limited to steps S111 to S115.

[0042] Step S111: monitoring and acquiring an initial data set, the initial data set comprising: an initial hyperspectral data set and initial inversion auxiliary data; wherein the inversion auxiliary data comprises: spatiotemporal dimension data and wavelength dimension data.

[0043] Step S112: Performing spatiotemporal analysis and segmentation processing, data index processing and data preprocessing on the initial spatiotemporal dimension data to obtain spatiotemporal dimension data divided by time series.

[0044] Step S113: performing spatiotemporal analysis segmentation processing and data indexing processing based on time attributes on the initial hyperspectral data set to obtain a hyperspectral data set divided in time sequence.

[0045] Step S114: performing band interval division processing and data index processing on the initial wavelength dimension data to obtain wavelength dimension data divided by band intervals.

[0046] Step S115: obtaining a target data set according to the hyperspectral data set divided in time series, the spatiotemporal dimension data divided in time series, and the wavelength dimension data divided in band intervals.

[0047] The inversion auxiliary data in step S111 and the method for obtaining the inversion auxiliary data are further described. In the process of inversion of atmospheric sulfur dioxide based on satellite remote sensing, inversion auxiliary data is needed; wherein, the inversion auxiliary data includes: spatiotemporal dimension data and wavelength dimension data; the spatiotemporal dimension data includes: atmospheric prior profile parameters, meteorological data, cloud height, cloud amount, and cloud pressure data. The wavelength dimension data includes: gas absorption cross-section data and surface reflectivity parameters.

[0048] Specifically, the atmospheric prior profile parameters include: sulfur dioxide prior profile, ozone prior profile, BrO prior profile, and HCHO prior profile. It should be noted that in the process of inverting the sulfur dioxide concentration in the atmosphere, in addition to sulfur dioxide, the gases that need to be considered to absorb light also include: ozone, BrO, HCHO and other interfering gases. The specific method of obtaining the atmospheric prior profile parameters is as follows: the daily global profile data of sulfur dioxide, BrO and HCHO can be simulated by the GEOS-Chem model to obtain the three input prior profiles required for this application: sulfur dioxide prior profile, BrO prior profile, and HCHO prior profile; wherein, the latitude and longitude spatial resolution of the sulfur dioxide prior profile, BrO prior profile, and HCHO prior profile is 2.5°×2.5°; in addition, the ozone prior profile is obtained by using the verified AURA MLS ozone product (MLS203.004); the ozone prior profile contains vertical stratification information of sampling in different latitude bands, and the latitude resolution of the ozone prior profile is about 5°.

[0049] Specifically, the meteorological data include: pressure data between the surface and the top of the troposphere, surface temperature data and temperature profile. The specific method of obtaining meteorological data is: obtaining the meteorological data required for this application through the daily reanalysis data set (FNL) jointly released by NCEP and NCAR; wherein, the latitude and longitude spatial resolution of the normalized meteorological data is 1°×1°.

[0050] Specifically, the cloud height, cloud amount and cloud pressure data are obtained by: obtaining the cloud height, cloud amount and cloud pressure data through the secondary cloud product data of the observed spectrum; the cloud height, cloud amount and cloud pressure data are input as fixed values.

[0051] Specifically, the gas absorption cross section data include: sulfur dioxide absorption cross section, ozone absorption cross section, BrO absorption cross section, HCHO absorption cross section. It is understandable that the gas absorption cross section intensity varies with wavelength. The specific method of obtaining the gas absorption cross section data is: first, through the existing measurement data, the initial sulfur dioxide absorption cross section, ozone absorption cross section, BrO absorption cross section, and HCHO absorption cross section are obtained; secondly, due to the large influence of ozone concentration on sulfur dioxide concentration, the sulfur dioxide absorption cross section and ozone absorption cross section also need to consider their changes with temperature; therefore, it is necessary to combine the temperature profile in the meteorological data to determine the final sulfur dioxide absorption cross section and ozone absorption cross section; among them, the sulfur dioxide absorption cross section considers the standard absorption cross sections at five temperatures of 203K, 223K, 243K, 273K and 293K; the ozone absorption cross section considers the standard absorption cross sections at four temperatures of 218K, 228K, 243K and 295K.

[0052] Specifically, the surface reflectance parameter is: the ratio of the ground observation spectrum and the solar observation spectrum at 347nm.

[0053] This application establishes a data segmentation strategy and a data spatiotemporal indexing strategy based on parameter spatiotemporal characteristics through steps S112 to S113.

[0054] Specifically, based on the inversion method of atmospheric sulfur dioxide concentration based on satellite observation provided in the embodiment of the present application, data preprocessing and spatiotemporal analysis segmentation processing need to be performed according to the necessary atmospheric prior profile parameters and meteorological data. The spatiotemporal analysis segmentation processing, data index processing and data preprocessing of the initial spatiotemporal dimension data in step S112 are further described.

[0055] The spatiotemporal analysis and segmentation of the initial spatiotemporal data specifically include: considering the small monthly changes of sulfur dioxide, HCHO and BrO, the monthly average profile data of the GEO-Chem simulation results are used as the input of the atmospheric prior profile parameters; considering the large daily changes of ozone and meteorological data, the daily AURAMLS ozone products and daily FNL meteorological data are still used. That is, the sulfur dioxide prior profile, BrO prior profile, and HCHO prior profile use the monthly average profile data of the EO-Chem simulation results; the ozone prior profile uses the daily average profile data, and the meteorological data uses the daily meteorological data.

[0056] The data preprocessing of the initial spatiotemporal data specifically includes: Since the vertical resolution of different data is different, data preprocessing is required to interpolate all the data to be processed to the same standard vertical distribution. The standard vertical distribution is set to 53 layers. Assuming that the surface is 1 standard atmosphere, the sampling step is about 0.015 standard atmosphere in the range of 1.0 to 0.6 standard atmosphere; in the range of 0.6 to 0 standard atmosphere, the sampling step is about 0.02 standard atmosphere. When preprocessing the atmospheric prior profile parameters, the volume ratio data of the vertical distribution of the gas is first converted into vertical column concentration data, and the vertical column concentration of the altitude layer is linearly interpolated to the standard vertical distribution based on the principle that the total amount of vertical column concentration is unchanged. When preprocessing the meteorological data, the meteorological field data in the standard vertical distribution can be directly obtained by linear interpolation.

[0057] The data indexing process of the initial spatiotemporal dimension data specifically includes: the data can be identified by using labels to facilitate indexing.

[0058] When establishing a spatiotemporal segmentation strategy for data, it is necessary to consider the multi-dimensional characteristics of the data, including the differences between different CCD instruments in satellite spectral data, the spatiotemporal distribution of prior gas profiles and meteorological data. The spatiotemporal analysis segmentation and data indexing processing of the initial hyperspectral data set in step S113 are further described, specifically including: according to the time attribute of the data set, the data set is segmented and sorted according to the monthly distribution, and an index is established for each month's data to facilitate rapid positioning and access.

[0059] In view of the characteristics of hyperspectral satellite data, in order to further parallelize the parameters related to the band in the inversion process, it is necessary to establish a spectral dimension segmentation strategy. The band interval division processing and data index processing of the initial wavelength dimension data in step S114 are further explained, specifically including: first, for the absorption characteristics of sulfur dioxide, its inversion window is selected as 311.0-326.0nm. Secondly, the band indicated by the inversion window is divided into several sub-bands, while ensuring that there is a certain overlapping area between every two adjacent bands to ensure the continuity and reliability of the data. Finally, for the input wavelength dimension data (specifically gas absorption cross-section data) that is independent of the interval distribution but related to the wavelength, the gas absorption cross-section data is divided according to the sub-band interval obtained by dividing the inversion window and a data index is established. Among them, the gas absorption cross-section of sulfur dioxide and ozone needs to be interpolated from the standard absorption cross-section to each vertical layer temperature of each pixel to obtain the temperature-related sulfur dioxide absorption cross-section and ozone absorption cross-section.

[0060] Through steps S111 to S115, a target data set can be obtained based on spatiotemporal analysis and processing, laying a data foundation for subsequent inversion calculations to obtain a target inversion result of sulfur dioxide concentration in the atmosphere.

[0061] Further explanation of step S130, based on the longitude and latitude information of the high-resolution satellite remote sensing spectrum, determining the CCD row and column range corresponding to the area to be inverted specifically includes: for the specific area that needs to be inverted, using the longitude and latitude information of the satellite observation spectrum to accurately locate the row and column range of the area to be inverted in the CCD instrument data; wherein, the row information of the row and column range reflects the latitude change, and the column information of the row and column range reflects the longitude change.

[0062] Further explanation of step S140: Based on the pre-processed spatiotemporal data, the spectral data of the CCD row and column range corresponding to the inversion area is subjected to spatial dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions.

[0063] The spatial dimension splitting process includes: taking into account the spatial resolution characteristics of the atmospheric prior profile parameters, the spectral data of the CCD row and column range corresponding to the observed area to be inverted are segmented according to the spatial distribution rate of the atmospheric prior profile parameters simulated by GEOS-Chem, which is 2.5°×2.5°. When the area to be segmented is mainly distributed along the latitude, it can be segmented according to column items to reduce the preprocessing time of the observed spectra of different column pixels.

[0064] The pixel marking process includes: after determining the row and column range of the area to be inverted in the CCD instrument data, the spectral data is marked by cloud height, cloud amount and cloud pressure data, where cloud amount is the percentage of cloud in the pixel to the pixel size. Pixels with cloud amount greater than 30% are marked as invalid data, and pixels with cloud amount less than 30% are marked as data to be calculated.

[0065] Further explanation of step S150: performing parallel data processing and spectral preprocessing on the multiple partitioned data blocks, performing spectral dimension splitting processing on the observed spectrum of each partitioned data block based on the wavelength dimension data, and obtaining multiple sub-band data blocks.

[0066] Step S150 is to perform parallel data processing and spectral preprocessing on multiple partitioned data blocks through multiple working nodes. It can be understood that, through step S140, the spectral data of the CCD row and column range corresponding to the area to be inverted is divided into multiple sub-areas to obtain multiple partitioned data blocks. In order to perform parallel processing of the spectral dimension in the subsequent step S160, the number of partitioned data blocks obtained by division should be much larger than the number of working nodes. For example, when M partitioned data blocks need to be processed in parallel, the main program will allocate N partitioned data blocks to N working nodes in parallel, and the remaining (MN) partitioned data blocks will be allocated to the working nodes in a cyclic manner, that is, they will be allocated one by one to the working node that is processed first. Among them, in the process of data processing for each partitioned data block, the invalid pixels marked as invalid in the partitioned data block will be skipped, and the invalid pixels will not be spectrally preprocessed; on the contrary, the valid pixels marked as valid in the partitioned data block are spectrally preprocessed to obtain partitioned data blocks including valid pixels after spectral preprocessing. Specifically, the spectrum preprocessing includes: spectrum wavelength calibration and spectrum intensity calibration using simulated spectrum data.

[0067] Further explanation of step S160: multiple sub-band data blocks are input into the radiation transfer model in parallel, and simulated parallel calculation processing is performed based on the simulated spectrum parameters to obtain the simulated spectrum corresponding to each sub-band data block. Among them, in the process of performing simulated parallel calculation processing, the radiation transfer model needs to combine the band information and other simulated spectrum parameters to simulate the observed spectrum to obtain the simulated spectrum. Specifically, the simulated spectrum parameters include: inversion auxiliary data and a preset simulated spectrum sampling rate. The inversion auxiliary data used here include but are not limited to: geometric information parameters, atmospheric condition parameters, cloud height, cloud amount, cloud pressure data, and reflectivity parameters. The geometric information parameters include: longitude and latitude, altitude, solar zenith angle, observation day fixed focus, etc., which come from the observed spectrum information; atmospheric condition parameters include: atmospheric prior profile parameters simulated by the GEOS-Chem model, meteorological data from NCEP reanalysis data; cloud height, cloud amount, cloud pressure data come from the secondary cloud product data of the observed spectrum; reflectivity parameters come from the ratio of the corresponding ground observation spectrum to the solar observation spectrum at 347nm. The simulated spectrum sampling rate is 0.2nm, and the simulated spectrum is also a normalized spectrum.

[0068] According to some embodiments of the present application, step S170 is further described, in which each simulated spectrum and observed spectrum are subjected to fitting processing and data integration processing to obtain an initial inversion result of sulfur dioxide concentration, including but not limited to steps S171 to S172.

[0069] Step S171: performing linear interpolation processing and optimization estimation processing on a simulated spectrum and an observed spectrum, and determining a sulfur dioxide concentration parameter that minimizes the difference between the simulated spectrum and the observed spectrum as a sub-inversion result within a sub-band interval;

[0070] Step S172: Integrate the sub-inversion results within multiple sub-band intervals to obtain an initial inversion result within the area to be inverted.

[0071] Step S171 specifically includes: using linear interpolation to interpolate the simulated spectrum to the wavelength distribution of the observed spectrum, thereby ensuring the consistency of the simulated spectrum and the observed spectrum in wavelength distribution.

[0072] Step S172 is specifically as follows: fitting the observed spectrum and the simulated spectrum within the inversion window, and inputting the obtained pixel's atmospheric sulfur dioxide concentration into the simulated spectrum parameters again, using Tikhonov regularization technology and embedding it into the Gauss-Newton iterative method, and minimizing the difference between the observed spectrum and the simulated spectrum through the optimization estimation method to obtain the final sulfur dioxide concentration.

[0073] Due to parallel processing, after fitting each simulated spectrum and observed spectrum, multiple partitioned inversion results of sulfur dioxide concentration will be obtained; they will be merged according to the row and column numbers of the original spectra to obtain the initial inversion results of sulfur dioxide concentration in the complete area.

[0074] Through step S171 to step S172, the initial inversion result in the area to be inverted can be obtained relatively quickly.

[0075] According to some embodiments of the present application, Figure 2 As shown, step S180 is further described, and the initial inversion result is sequentially subjected to stripe background correction processing and dimensional background correction processing to obtain a target inversion result of the corrected sulfur dioxide concentration, including but not limited to steps S181 to S184.

[0076] Step S181: Based on a preset fringe background correction value calculation formula, a first calculation process is performed on the initial inversion result to obtain a fringe background correction value;

[0077] Step S182: subtracting the fringe background correction value from the initial inversion result to obtain first inversion correction data of sulfur dioxide concentration after fringe background correction;

[0078] Step S183: Based on a preset latitude background correction value calculation formula, a second calculation process is performed on the first inversion correction data to obtain a latitude background correction value;

[0079] Step S184: subtract the latitude background correction value from the first inversion correction data to obtain the target inversion result after latitude background correction.

[0080] According to some embodiments of the present application, the stripe background correction value calculation formula is:

[0081] V t,k =MEDIAN[V k ];

[0082] Among them, V k represents the sulfur dioxide concentration distribution in the kth column of all the sulfur dioxide concentrations distributed at latitudes from -10° to 10°; MEDIAN represents the median function; V t,k It represents the fringe background correction value of the sulfur dioxide concentration distribution in the kth column.

[0083] According to some embodiments of the present application, the latitude background correction value calculation formula is:

[0084] V t,l =MEDIAN[V l ];

[0085] Among them, V lrepresents the sulfur dioxide concentration in the lth latitude band after stripe background correction; MEDIAN represents the median function; V t,l Represents the obtained latitude background correction value of the lth latitude band.

[0086] Through step S181 to step S184, a more accurate target inversion result after correction is obtained, thereby improving the accuracy of the inversion calculation of sulfur dioxide concentration.

[0087] Take an example to illustrate the stripe background correction processing and dimensional background correction processing in the remote sensing monitoring method of sulfur dioxide in the atmosphere of the present application.

[0088] First, select the clear and cloudless Pacific region corresponding to the observed spectrum within a month, and determine the row and column range of the area to be inverted in the CCD instrument data, where the selected latitude range needs to cover from -80° to 80°. According to the cloud cover data, mark the satellite data block pixels as invalid pixels and valid pixels. Note that the selected valid pixels must cover all column pixels. The initial inversion results of sulfur dioxide concentration in the complete area of ​​the clean background area are obtained by parallel computing.

[0089] Next, for the sulfur dioxide concentration distribution results within the latitude range of -10° to 10° in the initial inversion results, the median sulfur dioxide concentration of each column is calculated to obtain the sulfur dioxide concentration sequence changing along the column pixels, that is, the stripe background correction data of sulfur dioxide concentration (stripe background correction value), and the calculation formula is V t,k =MEDIAN[V k ]; where V k represents the sulfur dioxide concentration distribution in the kth column of all the sulfur dioxide concentrations distributed at latitudes from -10° to 10°; MEDIAN represents the median function; V t,k It represents the fringe background correction value of the sulfur dioxide concentration distribution in the kth column.

[0090] Then, within 10°, the initial inversion result is deducted from the fringe background correction value along each column to obtain the first inversion correction data of sulfur dioxide concentration after fringe background correction. The calculation formula is: V i,k ′=V i,k -V t,k ; Among them, V t,k V represents the fringe background correction value of the sulfur dioxide concentration distribution in the kth column; i,k represents the initial inversion result of sulfur dioxide concentration in row i and column j, V i,k ′ represents the first inversion correction data of sulfur dioxide concentration after fringe background correction.

[0091] Next, within the range of -80° to 80° of the first inversion correction data, the range of -80° to 80° is divided into different latitude bands with a step length of 2.5°, and the median of the first inversion correction data in different latitude bands is calculated to obtain the sulfur dioxide concentration sequence varying along the latitude, that is, the latitude background correction data of sulfur dioxide (that is, the latitude background correction value), and the calculation formula is: V t,l =MEDIAN[V l ]; where V l represents the sulfur dioxide concentration in the lth latitude band after stripe background correction; MEDIAN represents the median function; V t,l Represents the obtained latitude background correction value of the lth latitude band.

[0092] Finally, based on each dimensional band, the first inversion correction data of sulfur dioxide concentration after stripe background correction is subtracted from the latitude background correction value to obtain the target inversion result; the calculation formula is: V i,k ″=V i,k ′-V t,l ; Among them, V i,k ′ represents the first inversion correction data of sulfur dioxide concentration after fringe background correction in row i and column j, V t,l represents the latitude background correction value of the lth row, V i,k ″ represents the target inversion result of sulfur dioxide concentration obtained by the final correction.

[0093] According to some embodiments of the present application, after step S180, that is, after obtaining the target inversion result of the corrected sulfur dioxide concentration, the method also includes step S190: performing normalization processing on the target inversion result of the corrected sulfur dioxide concentration in parallel to obtain a regularly gridded regional sulfur dioxide concentration third-level data product.

[0094] Step S190 specifically includes: dividing the area to be inverted into several regular grids according to the required resolution to obtain gridded pixels; processing each gridded pixel using an asynchronous parallel computing method; during the parallel computing process, the calculation result of each gridded pixel is temporarily stored in the corresponding grid data structure to ensure the continuity and stability of the calculation process, and finally output to a file. When processing each gridded pixel, it is necessary to screen the sulfur dioxide concentration data of the original pixel to filter out those pixels with large fitting residuals; for the screened data, a weighted average method is used for processing based on pixel weights; wherein, the pixel weight is determined based on the inverse of the square of the distance from the pixel center to the grid pixel center, so as to more accurately reflect the contribution of different pixels to the final result.

[0095] like Figure 3As shown, the present invention also provides a remote sensing monitoring device for sulfur dioxide in the atmosphere, comprising:

[0096] The processor 301 may be implemented by a general-purpose central processing unit, a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0097] The memory 302 can be implemented in the form of a read-only memory, a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 302 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 302, and the processor 301 calls and executes the remote sensing method for sulfur dioxide in the atmosphere of the embodiment of the present application;

[0098] Input / output interface 303, used to implement information input and output;

[0099] The communication interface 304 is used to realize the communication interaction between the present apparatus and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0100] A bus 305 that transmits information between the various components of the device (e.g., the processor 301, the memory 302, the input / output interface 303, and the communication interface 304);

[0101] The processor 301 , the memory 302 , the input / output interface 303 and the communication interface 304 are connected to each other in communication within the device via the bus 305 .

[0102] An embodiment of the present application also provides an electronic device, including the remote sensing monitoring device for sulfur dioxide in the atmosphere as described above.

[0103] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned remote sensing monitoring method for sulfur dioxide in the atmosphere is implemented.

[0104] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are implemented to be located in one place, or may also be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0105] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0106] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present application. These equivalent deformations or substitutions are all included in the scope defined by the present application.

Claims

1. A remote sensing monitoring method for sulfur dioxide in the atmosphere, characterized in that: include: Perform spatiotemporal analysis on the initial data set obtained from monitoring to obtain the target data set; According to the time parameter, high-resolution satellite remote sensing spectrum, spatiotemporal dimension data and wavelength dimension data after data preprocessing are obtained from the target data set by indexing; Determine the CCD row and column range corresponding to the area to be inverted according to the latitude and longitude information of the high-resolution satellite remote sensing spectrum; According to the time-space dimension data after data preprocessing, the spectral data of the CCD row and column range corresponding to the inversion area is subjected to space dimension splitting processing and pixel labeling processing to obtain partition data blocks of multiple different partitions; Performing parallel data processing and spectral preprocessing on the multiple partitioned data blocks, performing spectral dimension splitting processing on the observed spectrum of each of the partitioned data blocks based on the wavelength dimension data, to obtain multiple sub-band data blocks; Inputting the plurality of sub-band data blocks into the radiation transfer model in parallel, performing simulation parallel calculation processing based on the simulation spectrum parameters, and obtaining a simulation spectrum corresponding to each of the sub-band data blocks; Performing fitting processing and data integration processing on each of the simulated spectra and the observed spectra to obtain an initial inversion result of sulfur dioxide concentration; The initial inversion result is subjected to stripe background correction processing and dimensional background correction processing in sequence to obtain a target inversion result of the corrected sulfur dioxide concentration.

2. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 1, characterized in that: The initial data set includes: an initial hyperspectral data set and initial inversion auxiliary data; the target data set includes: a hyperspectral data set divided by time sequence, spatiotemporal dimension data divided by time sequence, and wavelength dimension data divided by band intervals; The spatiotemporal analysis of the initial data set obtained by monitoring to obtain the target data set includes: Monitoring and acquiring the initial data set, the initial data set includes: the initial hyperspectral data set and the initial inversion auxiliary data; wherein the inversion auxiliary data includes: spatiotemporal dimension data and wavelength dimension data; Performing spatiotemporal analysis and segmentation processing, data index processing and data preprocessing on the initial spatiotemporal dimension data to obtain spatiotemporal dimension data divided by time series; Performing spatiotemporal analysis segmentation processing and data indexing processing based on time attributes on the initial hyperspectral data set to obtain a hyperspectral data set divided in time sequence; Performing band interval division processing and data index processing on the initial wavelength dimension data to obtain wavelength dimension data divided by band intervals; The target data set is obtained according to the hyperspectral data set divided in time sequence, the spatiotemporal dimension data divided in time sequence, and the wavelength dimension data divided in band intervals.

3. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 1, characterized in that: The fitting process and data integration process are performed on each of the simulated spectra and the observed spectra to obtain an initial inversion result of sulfur dioxide concentration, including: Performing linear interpolation processing and optimization estimation processing on the simulated spectrum and the observed spectrum, and determining the sulfur dioxide concentration parameter that minimizes the difference between the simulated spectrum and the observed spectrum as a sub-inversion result within a sub-band interval; The sub-inversion results within the multiple sub-band intervals are integrated to obtain the initial inversion result within the area to be inverted.

4. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 1, characterized in that: The initial inversion result is subjected to fringe background correction processing and dimensional background correction processing in sequence to obtain a target inversion result of the corrected sulfur dioxide concentration, including: Based on a preset fringe background correction value calculation formula, a first calculation process is performed on the initial inversion result to obtain a fringe background correction value; Subtract the fringe background correction value from the initial inversion result to obtain the first inversion correction data of sulfur dioxide concentration after fringe background correction; Based on a preset latitude background correction value calculation formula, a second calculation process is performed on the first inversion correction data to obtain a latitude background correction value; The latitude background correction value is subtracted from the first inversion correction data to obtain the target inversion result after latitude background correction.

5. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 4, characterized in that: The calculation formula of the stripe background correction value is: V t,k =MEDIAN[V k ]; Among them, V k represents the sulfur dioxide concentration distribution in the kth column of all the sulfur dioxide concentrations distributed at latitudes from -10° to 10°; MEDIAN represents the median function; V t,k It represents the fringe background correction value of the sulfur dioxide concentration distribution in the kth column.

6. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 4, characterized in that: The calculation formula of latitude background correction value is: V t,l =MEDIAN[V l ]; Among them, V l represents the sulfur dioxide concentration in the lth latitude band after stripe background correction; MEDIAN represents the median function; V t,l Represents the obtained latitude background correction value of the lth latitude band.

7. The remote sensing monitoring method for sulfur dioxide in the atmosphere according to claim 1, characterized in that: After obtaining the target inversion result of the corrected sulfur dioxide concentration, the method further comprises: The target inversion results of the corrected sulfur dioxide concentration are normalized in parallel to obtain a regularly gridded regional sulfur dioxide concentration level 3 data product.

8. A remote sensing monitoring device for sulfur dioxide in the atmosphere, characterized in that: It includes at least one processor and a memory for communicating with the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the remote sensing monitoring method for sulfur dioxide in the atmosphere as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It comprises the remote sensing monitoring device for sulfur dioxide in the atmosphere as described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the remote sensing monitoring method for sulfur dioxide in the atmosphere as described in any one of claims 1 to 7.

Images