A data fusion method and device based on satellite vertical observation products

Abstract

The present invention discloses a data fusion method and device based on satellite vertical observation products, comprising: respectively acquiring and screening first observation data of a first satellite product and second observation data of a second satellite product, and pairing the second observation data according to the time difference and space difference existing when observing a preset location to obtain second paired data; calculating a corresponding mapping weight matrix according to the first observation data and the second paired data, and mapping the first observation data and the second paired data according to a corresponding preset mapping relationship to obtain first mapping data and second mapping data; performing data joint calculation on the first mapping data and the second mapping data according to a preset data joint relationship to obtain pollutant vertical concentration data. The method is easy to expand and apply to joint algorithms of other pollutants, and solves the problems of high data processing difficulty, complex algorithms and high calculation costs.

Description

Technical Field

[0001] The present invention belongs to the field of satellite remote sensing technology application, and in particular relates to a data fusion method and device based on satellite vertical observation products. Background Art

[0002] Satellite remote sensing is a rapidly developing scientific field. Currently, remote sensing has formed a multi-level, multi-perspective and multi-field observation system that detects and monitors the world from the ground to the air and even to space, from information data collection, processing to interpretation, analysis and application, and has become an important means of obtaining information on earth resources and the environment. In the existing technology, most of the measurements of environmental pollutants are based on satellite L1-level data products, based on the optimal estimation theory, and by applying the forward model to combine the L1-level data products of two satellites, the joint vertical concentration is inverted.

[0003] However, the existing technology has the problem that the L1 data product is in the form of radiation spectrum, which is difficult to process. A forward model needs to be built in the process of combining data. The forward model needs to be specially developed when measuring different pollutants, which is time-consuming. In addition, the algorithm of the model is complex, which is not conducive to rapid data processing. The final data inversion process involves iteration and has high computational cost.

[0004] Therefore, those skilled in the art urgently need an algorithm that can reduce the difficulty of preliminary data processing, and that is easy to develop, has low computational cost, and can be easily expanded and applied to other pollutant combinations. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a data fusion method and device based on satellite vertical observation products, which effectively reduces the difficulty of early data processing compared with the existing technology, and is easy to expand and apply to joint algorithms for other pollutants.

[0006] In order to solve the above technical problems, the present invention provides a data fusion method based on satellite vertical observation products, which has the following characteristics:

[0007] Respectively obtaining and screening first observation data of a first satellite product at a preset location and second observation data of a second satellite product at the preset location, and pairing the second observation data according to a time difference and a space difference between the first satellite product and the second satellite product when observing the preset location, to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant;

[0008] According to the first observation data and the second paired data, respectively, corresponding mapping weight matrices are calculated, and according to the corresponding preset mapping relationship between the first observation data and the second paired data, the first observation data and the second paired data are mapped respectively to obtain first mapped data and second mapped data respectively;

[0009] According to a preset data joint relationship, data joint calculation is performed on the first mapping data and the second mapping data to obtain pollutant vertical concentration data.

[0010] Further, the obtaining and screening of the first observation data of the first satellite product at the preset location and the second observation data of the second satellite product at the preset location are specifically:

[0011] Based on the preset location, data of the first satellite product and the second satellite product at the preset location are respectively exported, and according to the standards in the satellite L2 level data product description file, the exported data of the first satellite product and the second satellite product at the preset location are respectively screened to obtain the first observation data and the second observation data, respectively.

[0012] Further, the second observation data is paired according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain the second paired data, specifically:

[0013] According to the time difference and spatial difference between the first satellite product and the second satellite product when observing the preset location, second observation data of the second satellite product within the error range of the time difference is selected, and among the selected second observation data, data with the smallest spatial difference of the second satellite product at the preset location is selected according to the spatial difference as second paired data of the second satellite product.

[0014] Furthermore, the corresponding mapping weight matrices are calculated respectively according to the first observation data and the second paired data, specifically:

[0015] The calculation of the mapping weight matrix is ​​divided into three cases. In the first case, when i=1,2,…,m-1; j=1,2,…,n;

[0016]

[0017]

[0018] In the second case, when j = 1, 2, ..., n;

[0019]

[0020] In the third case, when the mapping weight matrix The remaining unassigned elements When i=1, 2, ..., m; j=1, 2, ..., n;

[0021]

[0022] Where s is the corresponding first satellite product or the corresponding second satellite product, is the mapping weight matrix Elements of is the first observation data or the second paired data having m pressure layers, To preset the first mapping data or the second mapping data having n pressure layers, and They are and The pressure layers in descending order, and They are and elements.

[0023] Further, the first observation data and the second paired data are mapped according to the preset mapping relationship between the first observation data and the second paired data to obtain the first mapped data and the second mapped data, respectively, specifically:

[0024] The first observation data and the second paired data are mapped according to the corresponding mapping weight matrix and the mapping relationship in the preset pressure layer to obtain the preset corresponding first mapping data and second mapping data; wherein the preset mapping relationship is

[0025] Further, the performing data joint calculation on the first mapping data and the second mapping data according to the preset data joint relationship is specifically:

[0026] Substitute the first mapping data and the second mapping data into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is:

[0027]

[0028] Among them, x com is the vertical concentration data of pollutants, is the first mapping data, is the second mapping data, A s2 is the L2 level data averaging kernel of the second satellite product, I is the unit matrix, is the L2 data prior value of the second satellite product.

[0029] At the same time, the present invention also provides a data fusion device based on satellite vertical observation products, including: a data preprocessing module, a data mapping module and a data union module;

[0030] The data preprocessing module is used to respectively obtain and screen the first observation data of the first satellite product at the preset location and the second observation data of the second satellite product at the preset location, and pair the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant;

[0031] The data mapping module is used to calculate corresponding mapping weight matrices according to the first observation data and the second paired data, and map the first observation data and the second paired data according to the corresponding preset mapping relationship between the first observation data and the second paired data, to obtain first mapping data and second mapping data respectively;

[0032] The data union module is used to perform data union calculation on the first mapping data and the second mapping data according to a preset data union relationship to obtain pollutant vertical concentration data.

[0033] Further, the data preprocessing module is used to respectively obtain and filter the first observation data of the first satellite product at the preset location, and the second observation data of the second satellite product at the preset location, and pair the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain the second paired data, specifically:

[0034] Based on the preset location, data belonging to the first satellite product and the second satellite product are respectively exported, and according to the standard in the satellite L2 level data product description file, the exported data belonging to the first satellite product and the second satellite product are respectively screened to obtain first observation data and second observation data respectively; according to the time difference and spatial difference between the first satellite product and the second satellite product when observing the preset location, second observation data of the second satellite product within the error range of the time difference is selected, and among the selected second observation data, data with the smallest spatial difference of the second satellite product at the preset location is selected according to the spatial difference as second paired data of the second satellite product.

[0035] Further, the data mapping module is used to calculate corresponding mapping weight matrices according to the first observation data and the second paired data, and map the first observation data and the second paired data according to the corresponding preset mapping relationship between the first observation data and the second paired data, respectively, to obtain first mapping data and second mapping data, respectively, specifically:

[0036] The calculation of the mapping weight matrix is ​​divided into three cases. In the first case, when i=1,2,…,m-1; j=1,2,…,n;

[0037]

[0038]

[0039] In the second case, when j = 1, 2, ..., n;

[0040]

[0041] In the third case, when the mapping weight matrix The remaining unassigned elements When i=1, 2, ..., m; j=1, 2, ..., n;

[0042]

[0043] Where s is the corresponding first satellite product or the corresponding second satellite product, is the mapping weight matrix Elements of is the first observation data or the second paired data having m pressure layers, To preset the first mapping data or the second mapping data having n pressure layers, and They are and The pressure layers in descending order, and They are and Elements of

[0044] The first observation data and the second paired data are mapped according to the corresponding mapping weight matrix and the mapping relationship in the preset air pressure layer to obtain the preset corresponding first mapping data and second mapping data; wherein the mapping relationship of the preset air pressure layer is

[0045] Furthermore, the data union module is used to perform data union calculation on the first mapping data and the second mapping data according to a preset data union relationship to obtain pollutant vertical concentration data, specifically:

[0046] Substitute the first mapping data and the second mapping data into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is:

[0047]

[0048] Among them, x com is the vertical concentration data of pollutants, is the first mapping data, is the second mapping data, A s2 is the L2 level data averaging kernel of the second satellite product, I is the unit matrix, is the L2 data prior value of the second satellite product.

[0049] Implementing the embodiments of the present application has the following beneficial effects:

[0050] The present invention provides a data fusion method and device based on satellite vertical observation products. Compared with the prior art, the present invention avoids the difficulty in processing L1 data in the prior art by collecting and processing satellite L2 data, and maps and combines the processed L2 data, so that the present invention can be easily extended to other pollutant detection. The development is easy to implement and has low calculation cost and high accuracy, further improving the method for measuring pollutants in the application field of satellite remote sensing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flowchart of the steps of a data fusion method based on satellite vertical observation products provided by an embodiment of the present invention;

[0052] Figure 2 is a structural diagram of a data fusion device based on satellite vertical observation products provided by an embodiment of the present invention;

[0053] Figure 3 is the vertical concentration distribution of ozone pollutants in the embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be further described in detail below in conjunction with the drawings in the present invention. Obviously, the embodiments described here are only part of the embodiments of the present invention, not all of the embodiments, and are not intended to limit the present invention. Based on the embodiments in the present invention, other embodiments obtained by ordinary technicians in this field without making creative work are all within the scope of protection of the present invention.

[0055] See also Figure 1 , which is a data fusion method based on satellite vertical observation products provided by an embodiment of the present invention, comprising the following steps:

[0056] Step S101: respectively obtaining and screening first observation data of a first satellite product at a preset location and second observation data of a second satellite product at the preset location, and pairing the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant.

[0057] Preferably, based on the preset location, data of the first satellite product and the second satellite product at the preset location are respectively exported, and according to the standards in the satellite L2 level data product description file, the exported data of the first satellite product and the second satellite product at the preset location are respectively screened to obtain the first observation data and the second observation data, respectively.

[0058] In this embodiment, based on the preset location, the L2 ozone pollutant concentration data at the preset location belonging to the TES satellite product (TES L2 V8 ozone product) and the MLS satellite product (MLS L2 V5 ozone product) are exported respectively, and the L2 ozone pollutant concentration data are screened according to the quality mark in the L2 product manual and its corresponding quality standard; wherein the quality mark is an indicator for measuring the quality of ozone pollutant data, and the quality standard is the standard that the quality mark should meet.

[0059] In this embodiment, the TES satellite product is the first satellite product, the MLS satellite product is the second satellite product, and the satellite product L2 level data represents the volume mixing ratio of the data.

[0060] In this embodiment, preferably, the missing quality flag or data will be replaced by a filling value; wherein the filling value is generated by the satellite product according to the actual situation.

[0061] For MLSL2 satellite data product screening:

[0062] In the first step, the data with the quality flag as the fill value will not be eliminated, and then the observation sites will be screened: the observation sites with an odd "Status", the observation sites with a "Quality" less than or equal to 1, and the observation sites with a "Convergence" greater than or equal to 1.03 will be eliminated; in the second step, other screening: the data with an "O3Precision" less than or equal to 0 will be eliminated, and the data with a value less than 0 (including fill values ​​and negative values) will be eliminated;

[0063] For TESL Level 2 satellite data product screening:

[0064] In the first step, the quality flag is not eliminated when it is a fill value, and then the observation locations are screened: the observation locations with "SpeciesRetrievalQuality" not equal to 1 are eliminated, and the observation locations with "O3_Ccurve_QA" not equal to 1 are eliminated; the second step is other screening: data with a value less than 0 (including fill values ​​and negative values) are eliminated.

[0065] In this embodiment, the data after being filtered by the TESL2-level satellite data product is the first observation data, and the data after being filtered by the MLSL2-level satellite data product is the second observation data.

[0066] Furthermore, based on the time difference and spatial difference between the first satellite product and the second satellite product when observing the preset location, second observation data of the second satellite product within the error range of the time difference is selected, and among the selected second observation data, based on the spatial difference, data with the smallest spatial difference of the second satellite product at the preset location is selected as second paired data of the second satellite product.

[0067] In this embodiment, preferably, based on the time difference and spatial difference between the TES satellite product and the MLS satellite product when observing the preset location, the second observation data of the MLS satellite product within the time difference error range is selected, and the MLS satellite product data (i.e., the second observation data) that is 6-8 minutes ahead in time is selected, and among the selected second observation data, based on the spatial difference, the data with the smallest spatial difference of the MLS satellite product at the preset location is selected, that is, the data with the closest geographical distance to the TES satellite product data is selected as the pairing data of the MLS satellite product (the second pairing data of the second satellite product).

[0068] Step S102, calculating corresponding mapping weight matrices according to the first observation data and the second pairing data, and mapping the first observation data and the second pairing data according to the preset mapping relationship between the first observation data and the second pairing data, to obtain first mapping data and second mapping data respectively.

[0069] In this embodiment, the corresponding mapping weight matrices are calculated according to the observation data (first observation data) of the TES satellite product and the paired data (second paired data) of the MLS satellite product. There are three cases in total. In the first case, when i=1,2,…,m-1; j=1,2,…,n;

[0070]

[0071]

[0072] In the second case, when j = 1, 2, ..., n;

[0073]

[0074] In the third case, when the mapping weight matrix The remaining unassigned elements When i=1, 2, ..., m; j=1, 2, ..., n;

[0075]

[0076] Where s is the corresponding TES satellite product or the corresponding MLS satellite product, is the mapping weight matrix Elements of is the TES satellite product observation data (first observation data) or MLS satellite product paired data (second paired data) with m pressure layers, For presetting TES satellite product mapping data (first mapping data) or MLS satellite product mapping data (second mapping data) having n pressure layers, and They are and The pressure layers in descending order, and They are and elements.

[0077] In this embodiment, the observation data (first observation data) of the TES satellite product and the paired data (second paired data) of the MLS satellite product are mapped according to the corresponding mapping weight matrix and the mapping relationship in the preset pressure layer to obtain the preset corresponding mapping data of the TES satellite product (first mapping data) and the mapping data of the MLS satellite product (second mapping data); wherein the preset mapping relationship is

[0078] In this embodiment, the observation data (first observation data) of the TES satellite product is mapped through the mapping weight matrix of the TES satellite product and the MLS satellite product to obtain the TES satellite product mapping data (first mapping data), and the pairing data (second pairing data) of the MLS satellite product is mapped to obtain the MLS satellite product mapping data (second mapping data).

[0079] Preferably, in another embodiment, the mapping weight matrix preset on the pressure layer of the paired data (second paired data) of the MLS satellite product is calculated for the observation data (first observation data) of the TES satellite product, and mapping is performed to obtain the mapping data (first mapping data) of the TES satellite product on the pressure layer of the paired data (second paired data) of the MLS satellite product, that is, the paired data (second paired data) of the MLS satellite product in this embodiment is the mapping data (second mapping data) of the MLS satellite product. In this embodiment, only one mapping calculation of the observation data of the TES satellite product is required.

[0080] Step S103: performing data joint calculation on the first mapping data and the second mapping data according to a preset data joint relationship to obtain pollutant vertical concentration data.

[0081] Furthermore, the first mapping data and the second mapping data are substituted into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is:

[0082]

[0083] Among them, x com is the vertical concentration data of pollutants, is the first mapping data, is the second mapping data, A s2 is the L2 level data averaging kernel of the second satellite product, I is the unit matrix, is the L2 data prior value of the second satellite product.

[0084] In this embodiment, the TES satellite product mapping data and the MLS satellite product mapping data are substituted into the following data joint relationship to calculate the pollutant vertical concentration data. Figure 3 , is the vertical concentration distribution of pollutants:

[0085]

[0086] Among them, x com is the vertical concentration data of pollutants, Mapping data for TES satellite products, Mapping data for MLS satellite products, As2 is the L2 data averaging kernel of the MLS satellite product, I is the unit matrix, is the L2 data prior value of MLS satellite products.

[0087] Preferably, in another embodiment, the TES satellite product mapping data and the MLS satellite product mapping data are substituted into the data union relationship to calculate the pollutant vertical concentration data, and based on the calculated pollutant vertical concentration data, a mapping weight matrix of the pollutant vertical concentration data is calculated, and then based on the pollutant vertical concentration data mapping relationship of the preset pressure layer, the mapping data of the pollutant vertical concentration data in the preset pressure layer is mapped, so as to calculate the pollutant vertical concentration data of different pressure layers; wherein the pollutant vertical concentration data mapping relationship of the preset pressure layer is: It is the vertical concentration data of pollutants with m pressure layers after data joint calculation. is the mapping weight matrix of pollutant vertical concentration data, The vertical concentration data of pollutants with n pressure layers are preset after mapping.

[0088] Also, see Figure 2 The present invention also provides a data fusion device based on satellite vertical observation products, including: a data preprocessing module, a data mapping module and a data union module.

[0089] The data preprocessing module 201 is used to respectively obtain and screen the first observation data of the first satellite product at a preset location, and the second observation data of the second satellite product at the preset location, and pair the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant.

[0090] Furthermore, in this embodiment, based on the preset location, data belonging to the TES satellite product and the MLS satellite product are respectively exported, and according to the standards in the satellite L2 level data product description file, the exported data belonging to the TES satellite product and the MLS satellite product are respectively screened to obtain TES satellite product observation data and MLS satellite product observation data, respectively; according to the time difference and spatial difference between the TES satellite product and the MLS satellite product when observing the preset location, the MLS satellite product observation data within the error range of the time difference is selected, and among the selected MLS satellite product observation data, according to the spatial difference, the data with the smallest spatial difference of the MLS satellite product at the preset location is selected as the MLS satellite product pairing data.

[0091] The data mapping module 202 is used to calculate corresponding mapping weight matrices according to the first observation data and the second pairing data, and map the first observation data and the second pairing data according to the corresponding preset mapping relationship between the first observation data and the second pairing data, to obtain first mapping data and second mapping data, respectively.

[0092] Furthermore, in this embodiment, the calculation of the mapping weight matrix is ​​divided into three cases. In the first case, when i=1,2,…,m-1; j=1,2,…,n;

[0093]

[0094]

[0095] In the second case, when j=1,2,…,n;

[0096]

[0097] In the third case, when the mapping weight matrix The remaining unassigned elements When i=1,2,…,m;j=1,2,…,n;

[0098]

[0099] Where s is the corresponding TES satellite product or the corresponding MLS satellite product, is the mapping weight matrix Elements of It is the TES satellite product observation data with m-layer pressure layer or the MLS satellite product paired data. For pre-set TES satellite product mapping data or MLS satellite product mapping data with n pressure layers, and They are and The pressure layers in descending order, and They are and The observation data of the TES satellite product and the paired data of the MLS satellite product are mapped according to the corresponding mapping weight matrix and the mapping relationship in the preset pressure layer to obtain the mapping data of the TES satellite product and the mapping data of the MLS satellite product corresponding to the preset pressure layer; wherein the preset mapping relationship is

[0100] The data union module 203 is used to perform data union calculation on the first mapping data and the second mapping data according to a preset data union relationship to obtain pollutant vertical concentration data.

[0101] Furthermore, in this embodiment, the data union module is used to perform data union calculation on the TES satellite product mapping data and the MLS satellite product mapping data according to a preset data union relationship to obtain pollutant vertical concentration data, specifically:

[0102] Substitute the TES satellite product mapping data and the MLS satellite product mapping data into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is:

[0103]

[0104] Among them, x com is the vertical concentration data of pollutants, Mapping data for TES satellite products, Mapping data for MLS satellite products, A s2 is the L2 data averaging kernel of the MLS satellite product, I is the unit matrix, is the L2 data prior value of MLS satellite products.

[0105] Implementing the embodiments of the present invention has the following effects:

[0106] The embodiment of the present invention provides a data fusion method and device based on satellite vertical observation products, and proposes a method for detecting pollutants using satellite L2-level data products. The obtained L2-level product data is screened and eliminated, which reduces the difficulty of early data processing. The data is subsequently paired, mapped and combined, which can simplify the algorithm. The data fusion method based on satellite vertical observation products can be based on various pollutant data measured by satellite products, and can be easily expanded and applied to other pollutants for joint calculation, thereby reducing the cost of calculation and making pollutant detection development easy to implement.

[0107] The above are preferred embodiments of the present invention. It should be noted that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A data fusion method based on satellite vertical observation products, characterized in that: include: Respectively obtaining and screening first observation data of a first satellite product at a preset location and second observation data of a second satellite product at the preset location, and pairing the second observation data according to a time difference and a space difference between the first satellite product and the second satellite product when observing the preset location, to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant; According to the first observation data and the second paired data, the corresponding mapping weight matrices are calculated respectively, and the mapping is performed according to the mapping weight matrices corresponding to the first observation data and the second paired data and the mapping relationship in the preset pressure layer to obtain the preset corresponding first mapping data and second mapping data; wherein the preset mapping relationship is ; According to a preset data joint relationship, performing data joint calculation on the first mapping data and the second mapping data to obtain pollutant vertical concentration data; The calculation of the mapping weight matrix is ​​divided into three cases. In the first case, when , ; ； ； In the second case, when ; ； In the third case, when the mapping weight matrix The remaining unassigned elements hour, ； ； in, is the corresponding first satellite product or the corresponding second satellite product, is the mapping weight matrix Elements of is the first observation data or the second paired data having m pressure layers, To preset the first mapping data or the second mapping data having n pressure layers, and They are and The pressure layers in descending order, and They are and elements.

2. A data fusion method based on satellite vertical observation products as claimed in claim 1, characterized in that: The obtaining and screening of the first observation data of the first satellite product at the preset location and the second observation data of the second satellite product at the preset location respectively are specifically: Based on the preset location, data of the first satellite product and the second satellite product at the preset location are respectively exported, and according to the standards in the satellite L2 level data product description file, the exported data of the first satellite product and the second satellite product at the preset location are respectively screened to obtain the first observation data and the second observation data, respectively.

3. The data fusion method based on satellite vertical observation products according to claim 2, characterized in that: The pairing of the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain the second paired data is specifically: According to the time difference and spatial difference between the first satellite product and the second satellite product when observing the preset location, second observation data of the second satellite product within the error range of the time difference is selected, and among the selected second observation data, data with the smallest spatial difference of the second satellite product at the preset location is selected according to the spatial difference as second paired data of the second satellite product.

4. The data fusion method based on satellite vertical observation products according to claim 1, characterized in that: The performing data joint calculation on the first mapping data and the second mapping data according to the preset data joint relationship is specifically: Substitute the first mapping data and the second mapping data into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is: in, is the vertical concentration data of pollutants, is the first mapping data, is the second mapping data, is the L2 level data averaging kernel for the second satellite product, is the identity matrix, is the L2 data prior value of the second satellite product.

5. A data fusion device based on satellite vertical observation products, characterized in that: include: Data preprocessing module, data mapping module and data union module; The data preprocessing module is used to respectively obtain and screen the first observation data of the first satellite product at the preset location and the second observation data of the second satellite product at the preset location, and pair the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain second paired data; wherein the first observation data and the second observation data are L2 level data generated when the first satellite product and the second satellite product observe the same pollutant; The data mapping module is used to calculate the corresponding mapping weight matrix according to the first observation data and the second paired data, and perform mapping according to the mapping weight matrix corresponding to the first observation data and the second paired data and the mapping relationship in the preset pressure layer to obtain the preset corresponding first mapping data and second mapping data; wherein the preset mapping relationship is ; The data union module is used to perform data union calculation on the first mapping data and the second mapping data according to a preset data union relationship to obtain pollutant vertical concentration data; The calculation of the mapping weight matrix is ​​divided into three cases. In the first case, when , ; ； ； In the second case, when ; ； In the third case, when the mapping weight matrix The remaining unassigned elements hour, ； ； in, is the corresponding first satellite product or the corresponding second satellite product, is the mapping weight matrix Elements of is the first observation data or the second paired data having m pressure layers, To preset the first mapping data or the second mapping data having n pressure layers, and They are and The pressure layers in descending order, and They are and elements.

6. A data fusion device based on satellite vertical observation products as claimed in claim 5, characterized in that: The data preprocessing module is used to respectively obtain and filter the first observation data of the first satellite product at the preset location, and the second observation data of the second satellite product at the preset location, and pair the second observation data according to the time difference and space difference between the first satellite product and the second satellite product when observing the preset location to obtain the second paired data, specifically: Based on the preset location, data belonging to the first satellite product and the second satellite product are respectively exported, and according to the standard in the satellite L2 level data product description file, the exported data belonging to the first satellite product and the second satellite product are respectively screened to obtain first observation data and second observation data respectively; according to the time difference and spatial difference between the first satellite product and the second satellite product when observing the preset location, second observation data of the second satellite product within the error range of the time difference is selected, and among the selected second observation data, data with the smallest spatial difference of the second satellite product at the preset location is selected according to the spatial difference as second paired data of the second satellite product.

7. A data fusion device based on satellite vertical observation products as claimed in claim 6, characterized in that: The data union module is used to perform data union calculation on the first mapping data and the second mapping data according to a preset data union relationship to obtain pollutant vertical concentration data, specifically: Substitute the first mapping data and the second mapping data into the data joint relationship to calculate the pollutant vertical concentration data; wherein the data joint relationship is: in, is the vertical concentration data of pollutants, is the first mapping data, is the second mapping data, is the L2 level data averaging kernel for the second satellite product, is the identity matrix, is the L2 data prior value of the second satellite product.

Images