A method for generating spaceborne along-track microwave remote sensing datasets

Abstract

The present invention relates to a method for generating a satellite-borne along-track microwave remote sensing dataset. By overcoming the limitations of existing methods, the method dynamically adjusts the spatiotemporal traversal range during data matching, reduces the number of data comparisons, and can simply and efficiently perform quasi-synchronous spatiotemporal matching on multi-source remote sensing data to generate a remote sensing dataset. The method is a practical method for generating a dataset for along-track microwave remote sensing data.

Description

Technical Field

[0001] The invention belongs to the technical field of space microwave remote sensing. Background Art

[0002] Calibration and inversion of spaceborne microwave remote sensing data are core processes in quantitative microwave remote sensing Earth observation. This process requires the use of other satellite measurement data products or numerical model data products to correct data biases and verify inversion performance. In the process of batch processing microwave remote sensing data, facing massive amounts of remote sensing data, increasing the speed of remote sensing dataset generation can save a considerable amount of research time and is crucial to improving work efficiency.

[0003] The two existing methods for quasi-synchronous spatiotemporal matching of data are as follows:

[0004] 1) Conventional point-to-point comparison method

[0005] By setting certain temporal and spatial matching conditions, data that meets the conditions is retained to complete quasi-synchronous data matching. Due to the different satellite observation parameters and data storage methods, the matching of microwave remote sensing data of one track (or across days) requires traversing all orbital data of the second satellite on that day to determine whether the observation data of the two satellites overlap in time and space. For numerical model data, it has now developed into gridded data with a time resolution of 0.5 to 1 hour. According to the conventional point-to-point comparison method, all numerical model data within a day need to be traversed. It can be seen from this that the conventional method of matching two data sources has a huge amount of data processing and a long time required to generate data sets, which cannot meet the needs of high-efficiency batch processing of massive remote sensing data.

[0006] 2) Grid data matching method

[0007] This method divides the Earth's surface into uniform virtual grids at regular intervals of longitude and latitude. The data from two satellites are then interpolated or averaged onto the Earth's grid. Finally, the time difference between the two satellites' remote sensing observations corresponding to the same grid is calculated to see if it meets a set threshold. If so, the two satellites' data in that grid are considered synchronized, completing the matching process. Because this method uses longitude and latitude grids, the resulting matching results depend on the grid size. Furthermore, the data matrix size (rows and columns) differs from the matrix size (rows and columns) of the original along-track data, making it difficult to analyze and process the original along-track remote sensing data. Summary of the Invention

[0008] The technical problem solved by the present invention is: to overcome the shortcomings of the existing technology and provide a method for generating a satellite-borne along-track microwave remote sensing data set. By overcoming the limitations of the existing methods, the spatiotemporal traversal range during data matching is dynamically adjusted, the number of data comparisons is reduced, and the quasi-synchronous spatiotemporal matching of multi-source remote sensing data can be performed simply and efficiently to generate a remote sensing data set. The present invention is a practical method for generating a data set for along-track microwave remote sensing data.

[0009] The solution of the present invention is: a method for generating a satellite-borne along-track microwave remote sensing data set, comprising:

[0010] According to the observation time of along-track microwave remote sensing data called data 0, the time traversal range of data matching is determined;

[0011] Determine other source data files to be matched based on the time traversal range; the other source data include other source satellite along-track data to be synchronously matched, referred to as data 1, daily gridded satellite data, referred to as data 2, and gridded numerical model data with hourly time resolution, referred to as data 3;

[0012] Based on the longitude and latitude measurement information of along-track microwave remote sensing data, the spatial traversal range of each data unit is dynamically determined with N rows of data as a unit;

[0013] Perform quasi-synchronous spatiotemporal matching on the current data unit, read the data file of the data source to be matched, compare the latitude, longitude and time information of the microwave remote sensing data along the track in the unit and the data source to be matched point by point within the spatial traversal range of the current unit, and store the data after the matching is completed according to the quasi-synchronous spatiotemporal matching conditions;

[0014] Quasi-synchronous spatiotemporal matching is performed on other data units in turn until the generation of the full-track dataset is completed.

[0015] Preferably, the time traversal range for determining data matching includes:

[0016] Read the observation time information of data 0 and obtain the starting time t 0,start and the termination time t 0,end , and the corresponding date and

[0017] Considering the different time resolutions of various data sources, the time traversal range for data matching is set according to the measurement time of along-track data 0:

[0018] t range,1 =[t 0,start -tl,t 0,end ];

[0019]

[0020]

[0021] Where tl represents the measurement time length of along-track microwave remote sensing data; T step is the time resolution of data 3.

[0022] Preferably, determining other source data files to be matched includes:

[0023] Read the start time t of each track of data 1 1,start , search for t 1,start In t range,1 Data files in scope;

[0024] Read data date t2 of data 2 and search for t2 in t range,2 Data files in scope;

[0025] Rearrange and store data 3 according to its time resolution, name each data file according to its corresponding time, read each file time t3, search for t3 in t range,3 Data files within the scope.

[0026] Preferably, taking N rows of data as a unit, dynamically determining the spatial traversal range of each data unit includes:

[0027] Get the latitude lat0, longitude lon0 and time t0 corresponding to the current unit measurement data, and its corresponding minimum latitude lat min , maximum latitude lat max , minimum longitude lon min and maximum longitude lon max Information, and determine the spatial traversal range X of the unit when performing data matching:

[0028] The latitude range is: [lat min -2,lat max +2], in degrees;

[0029] The longitude range is: [lon min -2,lon max +2], the unit is degree.

[0030] Preferably, performing quasi-synchronous spatiotemporal matching on the data units includes:

[0031] Read the data time t of each other source data file to be matched in turn i and longitude and latitude information lon i ,lat i , where i represents the file index;

[0032] Determine other source data to be retained within the traversal range X of the current cell data space;

[0033] Set the time matching threshold t thre , spatial matching threshold (lat thre ,lon thre ), and compare the longitude and latitude information and time information of the along-track microwave remote sensing data and other retained source data in the unit point by point, and judge whether the quasi-synchronous matching conditions are met according to the matching threshold. If so, the matched data of the point are averaged and stored. Otherwise, it is considered that there is no matching data at this point, and the data of the point is set to an invalid value until the data matching of all points in the unit is completed.

[0034] Preferably, other source data within the traversal range X of the current unit data space are retained, and the corresponding data longitude and latitude meet the following conditions:

[0035] lon ix ={lon i |lon i ∈[lon min -2,lon max +2]}, the unit is degree;

[0036] lat ix = {lat i |lat i ∈[lat min -2,lat max +2]}, the unit is degree;

[0037] Where x represents the index of the current unit data, t ix Indicates the time information of retaining data;

[0038] If there is no data that meets the conditions in this unit, all points in this unit will have no matching data, all point values ​​will be set to invalid values, and the program will jump to the next unit.

[0039] Preferably, the quasi-synchronous matching conditions are as follows:

[0040] |lon ix -lon0|≤lon thre , and lat ix -lat0|≤lat thre , and |t ix -t0|≤t thre .

[0041] Preferably, the time matching threshold t thre The recommended setting is 25 minutes.

[0042] Preferably, the spatial matching threshold (lat thre ,lonthre ) is recommended to be set to half the spatial resolution of data 0.

[0043] The beneficial effects of the present invention compared with the prior art are:

[0044] (1) Based on the conventional point-to-point comparison method, the present invention reduces the number of file traversals for the data to be matched by setting a time traversal range, and dynamically adjusts the spatial traversal range of each unit by dividing the spatial data units according to the latitude and longitude information of the along-track microwave remote sensing data, thereby reducing the number of data comparisons for data matching and improving the matching efficiency.

[0045] (2) The present invention is a method for rapidly generating multi-source microwave remote sensing data sets in response to the needs of on-track microwave remote sensing data processing and quantitative inversion. The method is applicable to various forms of remote sensing data to be matched. The generated microwave remote sensing data set is consistent in size with the onboard on-track microwave remote sensing data matrix. It can be applied to the generation of massive remote sensing data sets and is simple to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Flowchart of the present invention;

[0047] Figure 2 Schematic diagram of dynamically adjusting the spatial traversal range. DETAILED DESCRIPTION

[0048] This paper proposes a method for rapidly generating multi-source spatiotemporally quasi-synchronous remote sensing datasets based on conventional point-to-point sequential comparison methods for satellite microwave remote sensing along-track data. The data source to be matched can be along-track or gridded data products at multiple temporal resolutions. This method has been validated using in-orbit measured data and boasts low computational complexity, simple operation, high operational efficiency, and strong practicality.

[0049] The present invention will be further described below in conjunction with the embodiments.

[0050] Satellite-borne microwave remote sensing measurement data products are mostly released in an along-track form. The data sources that can be used for quasi-synchronous matching include measurement data products from other satellites and numerical forecast model products. The data are mostly released in the form of along-track and global gridded data.

[0051] A method for generating satellite-borne along-track microwave remote sensing datasets is used for quasi-synchronous spatiotemporal matching of satellite-borne along-track microwave remote sensing data and other source data to generate remote sensing datasets. For convenience, the microwave remote sensing along-track data are referred to as data 0, the along-track data of other source satellites to be synchronously matched are referred to as data 1, the daily gridded satellite data are referred to as data 2, and the gridded numerical model data with hourly time resolution are referred to as data 3. Figure 1 As shown, the steps are as follows:

[0052] Step S1, determining the time traversal range of data matching based on the observation time of along-track microwave remote sensing data;

[0053] First, read the observation time information of data 0 and obtain the starting time t 0,start and the termination time t 0,end (specific to the second), and the corresponding date (specific to the day) and

[0054] Considering the different time resolutions of various data sources, the time traversal range for data matching is set according to the measurement time of along-track data 0:

[0055] For the data to be matched 1, t range,1 =[t 0,start -tl,t 0,end ];

[0056] For the data to be matched 2,

[0057] For the data to be matched 3,

[0058] Where tl represents the measurement time length of the on-track microwave remote sensing data, which is usually set to 2 hours based on the average working time of the geosynchronous orbit satellite; T step is the time resolution of the gridded numerical model data 3.

[0059] Step S2, determining other source data files and the number of files to be matched based on the time traversal range;

[0060] For data 1 to be matched, a data file usually contains the measurement data information of one track. Read the starting measurement time t of each track of data 1 1,start , search for t 1,start In t range,1 Data files within the scope.

[0061] For the data to be matched 2, usually a data file contains one day's data measurement information. Read the data file date t2 of the daily gridded satellite data product 2, and search for t2 in t range,2 Data files within the scope.

[0062] For the data to be matched 3: reorganize and store the gridded numerical model data 3 according to its time resolution, and name each data file according to its corresponding time (specifically to the hour). Read the time t3 of each file, and search for t3 in t range,3 Data files within the scope.

[0063] Step S3, see Figure 2As shown, based on the latitude and longitude measurement information of along-track microwave remote sensing data, N rows of data are taken as a unit, and the spatial traversal range of each data unit is dynamically determined;

[0064] For along-track microwave remote sensing data 0, take N lines of data as a unit, read the latitude lat0, longitude lon0 and time t0 of the unit data, and its corresponding minimum latitude lat min , maximum latitude lat max , minimum longitude lon min and maximum longitude lon max Information, determine the spatial traversal range X of the unit when performing data matching:

[0065] The latitude range is: [lat min -2,lat max +2], in degrees;

[0066] The longitude range is: [lon min -2,lon max +2], the unit is degree.

[0067] Step S4: Read the data file of the data source to be matched, compare the latitude and longitude information and time information of the along-track microwave remote sensing data in each unit with the data source to be matched point by point within the spatial traversal range of each unit, determine whether the quasi-synchronous spatiotemporal matching conditions are met, and store the data;

[0068] Read the data time t of each other data source to be matched file in turn i and longitude and latitude information lon i ,lat i , where i represents the file index.

[0069] Keep other source data within the unit data space traversal range X, the corresponding longitude lon ix and latitude lat ix The following conditions are met:

[0070] lon ix ={lon i |lon i ∈[lon min -2,lon max +2]}, the unit is degree.

[0071] lat ix = {lat i |lat i ∈[lat min -2,lat max +2]}, the unit is degree.

[0072] Where x represents the index of the cell data (N rows), and t is used ix Indicates the time information of the retained data. If there is no data that meets the conditions in this unit, the matching data of all points in this unit will be set to -999 and jump to the next unit.

[0073] Set the time matching threshold t thre (25 minutes is recommended), spatial matching threshold (lat thre ,lon thre ) (recommended to be half of the spatial resolution of data 0), and compare the latitude, longitude, and time information of the along-track microwave remote sensing data in the unit with the retained other source data point by point. According to the matching threshold, it is determined whether the quasi-synchronous matching conditions are met. If so, the matched data are averaged and stored. Otherwise, the data of the point is set to -999 until the data matching of all points in the unit is completed. The quasi-synchronous matching conditions are as follows:

[0074] |lon ix -lon0|≤lon thre , and |lat ix -lat0|≤lat thre , and |t ix -t0|≤t thre .

[0075] Step S5: Perform quasi-synchronous spatiotemporal matching on each data unit in turn until the generation of the full track dataset is completed.

[0076] Repeat steps S3 and S4 for the next data unit in turn to complete the generation of the full track dataset.

[0077] The present invention provides a method for generating a satellite-borne, along-track microwave remote sensing dataset. This method addresses the massive volume of satellite-borne microwave remote sensing data and the growing demand for rapid remote sensing dataset generation. This method dynamically adjusts the temporal and spatial traversal ranges of data matching, while ensuring the dataset size is consistent with the original along-track remote sensing data matrix. By limiting the temporal traversal range and dynamically adjusting the spatial traversal range, the method significantly reduces the number of remote sensing data comparisons and significantly improves data matching efficiency. This method is applicable to a variety of remote sensing data formats, is simple to implement, and highly practical, making it widely applicable to the field of satellite-borne microwave remote sensing data processing.

[0078] Although the present invention has been disclosed above in terms of preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solutions of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the scope of protection of the technical solutions of the present invention.

[0079] Parts of the present invention that are not described in detail belong to the common knowledge of those skilled in the art.

Claims

1. A method for generating a spaceborne along-track microwave remote sensing dataset, characterized in that include: According to the observation time of along-track microwave remote sensing data called data 0, the time traversal range of data matching is determined; Determine other source data files to be matched based on the time traversal range; the other source data include other source satellite along-track data to be synchronously matched, referred to as data 1, daily gridded satellite data, referred to as data 2, and gridded numerical model data with hourly time resolution, referred to as data 3; Based on the longitude and latitude measurement information of along-track microwave remote sensing data, the spatial traversal range of each data unit is dynamically determined with N rows of data as a unit; Perform quasi-synchronous spatiotemporal matching on the current data unit, read the data file of the data source to be matched, compare the latitude, longitude and time information of the along-track microwave remote sensing data in the unit and the data source to be matched point by point within the spatial traversal range of the current unit, and store the data after matching according to the quasi-synchronous spatiotemporal matching conditions; Quasi-synchronous spatiotemporal matching is performed on other data units in turn until the generation of the full-track dataset is completed.

2. The method according to claim 1, wherein: The time traversal range for determining data matching includes: Read the observation time information of data 0 and obtain the starting time t 0,start and the end time t 0,end , and the corresponding date and Considering the different time resolutions of various data sources, the time traversal range for data matching is set according to the measurement time of along-track data 0: t range,1 =[t 0,start -tl,t 0,end ]: Where tl represents the measurement time length of along-track microwave remote sensing data; T step is the time resolution of data 3.

3. The method according to claim 2, wherein: Other source data files to be matched include: Read the start time t of each track of data 1 1,start , search for t 1,start In t range,1 Data files in scope; Read data date t2 of data 2 and search for t2 in t range,2 Data files in scope; Rearrange and store data 3 according to its time resolution, name each data file according to its corresponding time, read each file time t3, search for t3 in t range,3 Data files within the scope.

4. The method according to claim 1, wherein: Taking N rows of data as a unit, dynamically determining the spatial traversal range of each data unit includes: Get the latitude lat0, longitude lon0 and time t0 corresponding to the current unit measurement data, and its corresponding minimum latitude lat min , maximum latitude lat max , minimum longitude lon min and maximum longitude lon max Information, and determine the spatial traversal range X of the unit when performing data matching: The latitude range is: [lat min -2,lat max +2], in degrees; The longitude range is: [lon min -2,lon max +2], the unit is degree.

5. The method according to claim 4, characterized in that: Quasi-synchronous spatiotemporal matching of data units includes: Read the data time t of each other source data file to be matched in turn i and longitude and latitude information lon i ,lat i , where i represents the file index; Determine other source data to be retained within the traversal range X of the current cell data space; Set the time matching threshold t thre , spatial matching threshold lat thre ,lon thre , and compare the longitude and latitude information and time information of the along-track microwave remote sensing data and other retained source data in the unit point by point, and judge whether the quasi-synchronous matching conditions are met according to the matching threshold. If so, the matched data of the point are averaged and stored, otherwise it is considered that there is no matching data at this point, and the data of the point is set to an invalid value until the data matching of all points in the unit is completed.

6. The method according to claim 5, characterized in that: Keep other source data within the traversal range X of the current cell data space, and the corresponding data longitude and latitude meet the following conditions: lon ix ={lon i |lon i ∈[lon min -2,lon max +2]}, the unit is degree; lat ix = {lat i |lat i ∈[lat min -2,lat max +2]}, the unit is degree; Where x represents the index of the current unit data, t ix Indicates the time information of retaining data; If there is no data that meets the conditions in this unit, all points in this unit will have no matching data, all point values ​​will be set to invalid values, and the program will jump to the next unit.

7. The method according to claim 6, characterized in that: The quasi-synchronous matching conditions are as follows: |lon ix -lon0|≤lon thre , and |lat ix -lat0|≤lat thre , and |t ix -t0|≤t thre .

8. The method according to claim 7, wherein: Time matching threshold t thre The setting is 25 minutes.

9. The method according to claim 7, wherein: Spatial matching threshold lat thre ,lon thre is set to half the spatial resolution of data 0.

Images