Global Precipitation Data Mosaic Synthesis Method and Device

Abstract

The present invention provides a method and device for splicing and synthesizing global precipitation data, relating to the technical field of data processing, and comprising the following steps: performing spatial interpolation processing on the precipitation data of ground observations at global meteorological stations to obtain rasterized precipitation data of station observations; acquiring low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the precipitation data of each station observation; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the precipitation data of station observations; calibrating the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the precipitation data of station observations according to the precipitation data of station observations to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data; splicing and synthesizing the calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data to obtain spatially continuous global precipitation data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and device for splicing and synthesizing global precipitation data. Background Art

[0002] Precipitation is a key element in the global water cycle process and is of great significance for research and applications in aspects such as climate, hydrology, ecology, and disasters.

[0003] The acquisition methods of precipitation data mainly include station observations, satellite remote sensing data, etc. Although the station observation data has high accuracy, it cannot effectively reflect the spatial distribution characteristics of precipitation in areas with sparse station distributions. Satellite remote sensing technology has significant application potential in obtaining long-term and large-scale regional precipitation, and can effectively reflect the continuity and heterogeneity characteristics of precipitation spatio-temporal distribution. However, its spatial distribution range is usually limited to the mid-low latitude regions between 50° north and south latitudes or between 60° north and south latitudes.

[0004] Therefore, how to effectively obtain comprehensive global precipitation data has become an urgent problem to be solved in the industry. Summary of the Invention

[0005] The present invention provides a method and device for splicing and synthesizing global precipitation data to solve the problem of how to effectively obtain comprehensive global precipitation data in the prior art.

[0006] The present invention provides a method for splicing and synthesizing global precipitation data, including the following steps:

[0007] Perform spatial interpolation processing on the station ground observation precipitation data of global meteorological stations to obtain rasterized station observation precipitation data;

[0008] Obtain low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the station observation precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the station observation precipitation data;

[0009] Calibrate the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the station observation precipitation data according to the station observation precipitation data to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0010] Splice and synthesize the calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data to obtain spatially continuous global precipitation data.

[0011] A method for splicing and synthesizing global precipitation data provided by the present invention calibrates low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site-observed precipitation data according to the spatial resolution of the site-observed precipitation data, including:

[0012] Analyze the site-observed precipitation data through a precipitation remote sensing calibration model to determine the expected low-latitude precipitation remote sensing data; wherein, the precipitation remote sensing calibration model is constructed based on the historical site-observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the site-observed precipitation data;

[0013] Calibrate the low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain the calibrated low-latitude precipitation remote sensing data;

[0014] Analyze the site-observed precipitation data through a precipitation reanalysis calibration model to determine the expected high-latitude precipitation reanalysis data; wherein, the precipitation reanalysis calibration model is constructed based on the historical site-observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the site-observed precipitation data;

[0015] Calibrate the high-latitude precipitation reanalysis data according to the expected high-latitude precipitation reanalysis data to obtain the calibrated high-latitude precipitation reanalysis data.

[0016] According to a method for splicing and synthesizing global precipitation data provided by the present invention, the analyzing the site-observed precipitation data through a precipitation remote sensing calibration model to determine the expected low-latitude precipitation remote sensing data includes:

[0017] Obtain the historical site-observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the grid points of the site-observed precipitation data;

[0018] Calculate the first cumulative probability density value of the historical site-observed precipitation data of the grid point, and obtain the historical low-latitude precipitation remote sensing data closest to the first cumulative probability density value among each historical low-latitude precipitation remote sensing data as the expected low-latitude precipitation remote sensing data.

[0019] According to a method for splicing and synthesizing global precipitation data provided by the present invention, the analyzing the site-observed precipitation data through a precipitation reanalysis calibration model to determine the expected high-latitude precipitation reanalysis data includes:

[0020] Obtain the historical site-observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the grid points of the site-observed precipitation data;

[0021] Calculate the first cumulative probability density value of the historical station-observed precipitation data at the grid lattice points, and obtain the historical high-latitude precipitation reanalysis data closest to the first cumulative probability density value among the various historical high-latitude precipitation reanalysis data as the expected high-latitude precipitation reanalysis data.

[0022] According to a global precipitation data stitching and synthesis method provided by the present invention, perform spatial interpolation processing on the station ground-observed precipitation data of global meteorological stations to obtain rasterized station-observed precipitation data, including:

[0023] Perform data cleaning, outlier detection, and missing value filling on the station ground-observed precipitation data of the global meteorological stations to achieve quality control and obtain the quality-controlled station ground-observed precipitation data;

[0024] Based on a preset interpolation method, interpolate the quality-controlled station ground-observed precipitation data into a defined grid system to obtain rasterized station-observed precipitation data; wherein, the preset interpolation method includes at least one of the following: Kriging method, inverse distance weighting, and spline interpolation.

[0025] According to a global precipitation data stitching and synthesis method provided by the present invention, stitch and synthesize the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain spatially continuous global precipitation data, including:

[0026] Obtain the calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data for each grid lattice point; wherein, the grid lattice points are the spatial lattice points corresponding to the rasterized station-observed precipitation data;

[0027] Based on the calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data of each grid lattice point, perform stitching and synthesis to obtain spatially continuous global precipitation data.

[0028] The present invention also provides a global precipitation data stitching and synthesis device, including:

[0029] A processing module for performing spatial interpolation processing on the station ground-observed precipitation data of global meteorological stations to obtain rasterized station-observed precipitation data;

[0030] A first analysis module for obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the respective station-observed precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the station-observed precipitation data;

[0031] A second analysis module, configured to calibrate low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site-observed precipitation data according to the site-observed precipitation data, to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0032] A synthesis module, configured to splice and synthesize the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution.

[0033] The present invention further provides a globally continuous precipitation data splicing and synthesizing device, and the device is further configured to:

[0034] Analyze the site-observed precipitation data through a precipitation remote sensing calibration model to determine expected low-latitude precipitation remote sensing data; wherein, the precipitation remote sensing calibration model is constructed according to historical site-observed precipitation data corresponding to the site-observed precipitation data and historical low-latitude precipitation remote sensing data;

[0035] Calibrate the low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain calibrated low-latitude precipitation remote sensing data;

[0036] Analyze the site-observed precipitation data through a precipitation reanalysis calibration model to determine expected high-latitude precipitation reanalysis data; wherein, the precipitation reanalysis calibration model is constructed according to historical site-observed precipitation data corresponding to the site-observed precipitation data and historical high-latitude precipitation reanalysis data;

[0037] Calibrate the high-latitude precipitation reanalysis data according to the expected high-latitude precipitation reanalysis data to obtain calibrated high-latitude precipitation reanalysis data.

[0038] The present invention further provides a globally continuous precipitation data splicing and synthesizing device, and the device is further configured to:

[0039] Obtain historical site-observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the grid points of the site-observed precipitation data;

[0040] Calculate a first cumulative probability density value of the historical site-observed precipitation data of the grid point, and obtain the historical low-latitude precipitation remote sensing data that is closest to the first cumulative probability density value among each historical low-latitude precipitation remote sensing data as the expected low-latitude precipitation remote sensing data.

[0041] The present invention further provides a globally continuous precipitation data splicing and synthesizing device, and the device is further configured to:

[0042] Obtain historical site-observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the grid points of the site-observed precipitation data;

[0043] Calculate the first cumulative probability density value of the historical site-observed precipitation data of the grid lattice points, and obtain the historical high-latitude precipitation reanalysis data closest to the first cumulative probability density value among each historical high-latitude precipitation reanalysis data as the expected high-latitude precipitation reanalysis data.

[0044] The present invention also provides a global precipitation data splicing and synthesizing device, and the device is further configured to:

[0045] Perform data cleaning, outlier detection, and missing value filling on the site surface-observed precipitation data of the global meteorological sites to achieve quality control, and obtain the quality-controlled site surface-observed precipitation data.

[0046] Interpolate the quality-controlled site surface-observed precipitation data into a defined grid system based on a preset interpolation method to obtain rasterized site-observed precipitation data; wherein, the preset interpolation method includes at least one of the following: Kriging method, inverse distance weighting, and spline interpolation.

[0047] The present invention also provides a global precipitation data splicing and synthesizing device, and the device is further configured to:

[0048] Obtain the calibrated low-latitude precipitation remote sensing data or calibrated high-latitude precipitation reanalysis data of each grid lattice point; wherein, the grid lattice point is the spatial lattice point corresponding to the rasterized site-observed precipitation data.

[0049] Perform splicing and synthesis based on the calibrated low-latitude precipitation remote sensing data or calibrated high-latitude precipitation reanalysis data of each grid lattice point to obtain spatially continuous global precipitation data.

[0050] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the global precipitation data splicing and synthesizing method as described in any one of the above.

[0051] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the global precipitation data splicing and synthesizing method as described in any one of the above.

[0052] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the global precipitation data splicing and synthesizing method as described in any one of the above.

[0053] The global precipitation data mosaicking and synthesizing method and device provided by the present invention perform spatial interpolation processing on the ground observation precipitation data of global meteorological stations to obtain rasterized station observation precipitation data, which can effectively reduce the spatial sampling error of station observations and make the spatial resolution of station observation data consistent with that of precipitation remote sensing data and precipitation reanalysis data. By obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the precipitation data observed at each station. This step ensures the consistency of spatial resolution between different data sources and provides a basis for subsequent calibration and mosaicking. According to the rasterized station observation precipitation data, the matched low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are calibrated to obtain the calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data. This step improves the data accuracy of precipitation remote sensing data and precipitation reanalysis data and the consistency of their spatio-temporal distribution by establishing statistical relationships and calibration models. The calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are mosaicked and synthesized to obtain globally continuous precipitation data. It solves the problem of inconsistent spatio-temporal distribution in the area near the mosaic line at 50° north or south latitude or 60° north or south latitude, and realizes global-scale precipitation analysis and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0055] Figure 1 It is a schematic flowchart of the global precipitation data mosaicking and synthesizing method provided by the present invention;

[0056] Figure 2 It is a schematic flowchart of the global precipitation data mosaicking and synthesizing process provided by the present invention;

[0057] Figure 3 It is a schematic structural diagram of the global precipitation data mosaicking and synthesizing device provided by the present invention;

[0058] Figure 4 It is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the protection scope of the present invention.

[0060] Figure 1 is a schematic flowchart of the method for splicing and synthesizing global precipitation data provided by the present invention. As Figure 1 shown, the method includes the following:

[0061] Step 110: Perform spatial interpolation processing on the site ground observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data;

[0062] In the present invention, before performing spatial interpolation, it is necessary to perform quality control on the collected site observation precipitation data to ensure the accuracy and reliability of the data.

[0063] Adopt appropriate spatial interpolation methods, such as optimal interpolation method, thin plate spline interpolation method, etc., to process the site observation precipitation data. These methods can consider the influence of precipitation spatial structure, so as to more accurately reflect the spatial distribution characteristics of precipitation.

[0064] Convert the interpolated data into a rasterized format, that is, each data point corresponds to a specific spatial position to form raster data. The purpose of this step is to make the spatial resolution of the site observation data consistent with the spatial resolution of precipitation remote sensing data and precipitation reanalysis data, such as 0.25°.

[0065] Through the rasterized site observation precipitation data, the spatial sampling error of site observation can be effectively reduced, and the spatial continuity and coverage of the data can be improved.

[0066] The rasterized site observation precipitation data will be used in the subsequent calibration process to improve the data accuracy and the consistency of spatio-temporal distribution of precipitation remote sensing data in low-latitude regions and precipitation reanalysis data in high-latitude regions.

[0067] Step 120: Obtain low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the site observation precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the site observation precipitation data;

[0068] In the embodiments of the present invention, when obtaining low-latitude precipitation remote sensing data (such as TRMM, CHIRPS, etc.) and high-latitude precipitation reanalysis data (such as ERA5), it is necessary to ensure that the spatial resolution of these data matches the spatial resolution of the in-situ observed precipitation data. For example, if the spatial resolution of the in-situ observed data is 0.25°, remote sensing and reanalysis data of the same resolution need to be obtained.

[0069] ERA5 provides global hourly high-temporal-resolution surface and atmospheric variable information with a spatial resolution of 0.25°×0.25°, which is suitable as a precipitation reanalysis data source for high-latitude regions. The ERA5-Land dataset provides a finer spatial resolution of 0.1°×0.1°, which is suitable for the analysis of land precipitation data.

[0070] CMORPH is a precipitation data product between 60° north and south latitudes. Its original temporal resolution is 30 minutes and the spatial resolution is 8 km×8 km. It can be further processed into daily precipitation grid products with a resolution of 0.25°×0.25° and hourly precipitation grid products with a resolution of 0.1°×0.1°, which is suitable as a low-latitude precipitation remote sensing data source.

[0071] Through the above data sources, low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data with the same spatial resolution as the in-situ observed precipitation data can be obtained, ensuring the matching and consistency between the data, and providing an accurate data basis for subsequent calibration and mosaicking synthesis.

[0072] Step 130: Calibrate the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data that match the spatial resolution of the in-situ observed precipitation data according to the in-situ observed precipitation data, to obtain the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data;

[0073] In the present invention, a calibration model is established using the statistical relationship between the matched ground observation data and precipitation remote sensing data. This can be accomplished by various methods, including linear regression, probability density matching methods, or machine learning methods, etc.

[0074] In the present invention, for low-latitude precipitation remote sensing data (such as TRMM, CHIRPS), at each grid point, the matched ground observation data and precipitation remote sensing data within an appropriate spatio-temporal window are collected, and a calibration model is established. For example, when using the probability density matching method, matching data are collected for each precipitation remote sensing data grid point to be calibrated, and the cumulative probability density value is calculated, and the precipitation remote sensing data is calibrated according to the deviation.

[0075] For high-latitude precipitation reanalysis data (such as ERA5), at each grid point, appropriate ground observation data and precipitation reanalysis data within a suitable spatio-temporal window are collected, and a calibration model is established. The same method as for low-latitude data is used to calibrate the reanalysis data.

[0076] Through calibration, the data accuracy of precipitation remote sensing data and precipitation reanalysis data can be improved, the uncertainty existing in the data can be reduced, and the spatio-temporal distribution consistency between precipitation remote sensing data and precipitation reanalysis data can be enhanced.

[0077] Step 140: Stitch and synthesize the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in terms of spatial distribution.

[0078] In the present invention, before stitching, it is ensured that both the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data have been calibrated according to the precipitation data observed at stations, so as to ensure their data accuracy and spatio-temporal distribution consistency.

[0079] Align the calibrated low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data in the stitching area (the area near the stitching line at 50°N / S or 60°N / S), ensuring that the two data sets can smoothly transition at the boundary of the stitching area and reducing discontinuities. Use appropriate fusion techniques, such as weighted average, regression methods, machine learning algorithms, etc., to merge the two data sets into a continuous global precipitation data set. The choice of fusion technique depends on the characteristics of the data and the expected application requirements.

[0080] In the present invention, by performing spatial interpolation on the ground observation precipitation data of global meteorological stations, rasterized station observation precipitation data are obtained, which can effectively reduce the spatial sampling error of station observations and make the spatial resolution of station observation data consistent with that of precipitation remote sensing data and precipitation reanalysis data. By obtaining the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data that match each station observation precipitation data. This step ensures the spatial resolution consistency between different data sources and provides a basis for subsequent calibration and stitching. According to the rasterized station observation precipitation data, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are calibrated to obtain the calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data. This step improves the data accuracy of precipitation remote sensing data and precipitation reanalysis data and their spatio-temporal distribution consistency by establishing statistical relationships and calibration models. The calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are stitched and synthesized to obtain globally continuous precipitation data in terms of spatial distribution. The problem of inconsistent spatio-temporal distribution in the area near the stitching line at 50°N / S (or 60°N / S) is solved, and global-scale precipitation analysis applications are realized.

[0081] Optionally, according to the spatial resolution of the precipitation data observed at the stations, calibrate the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data that match the spatial resolution of the precipitation data observed at the stations, including:

[0082] Analyze the precipitation data observed at the stations through a precipitation remote sensing calibration model to determine the expected low-latitude precipitation remote sensing data; wherein, the precipitation remote sensing calibration model is constructed based on the historical precipitation data observed at the stations corresponding to the precipitation data observed at the stations and the historical low-latitude precipitation remote sensing data;

[0083] Calibrate the low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain the calibrated low-latitude precipitation remote sensing data;

[0084] Analyze the precipitation data observed at the stations through a precipitation reanalysis calibration model to determine the expected high-latitude precipitation reanalysis data; wherein, the precipitation reanalysis calibration model is constructed based on the historical precipitation data observed at the stations corresponding to the precipitation data observed at the stations and the historical high-latitude precipitation reanalysis data;

[0085] Calibrate the high-latitude precipitation reanalysis data according to the expected high-latitude precipitation reanalysis data to obtain the calibrated high-latitude precipitation reanalysis data.

[0086] In the present invention, a precipitation remote sensing calibration model is constructed by using the historical precipitation data observed at the stations corresponding to the precipitation data observed at the stations and the historical low-latitude precipitation remote sensing data. This model can analyze the precipitation data observed at the stations and determine the expected low-latitude precipitation remote sensing data.

[0087] Calibrate the actual low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain the calibrated low-latitude precipitation remote sensing data. This step ensures the accuracy and reliability of the remote sensing data.

[0088] Similarly, a precipitation reanalysis calibration model is constructed by using the historical precipitation data observed at the stations corresponding to the precipitation data observed at the stations and the historical high-latitude precipitation reanalysis data. This model is used to analyze the precipitation data observed at the stations and determine the expected high-latitude precipitation reanalysis data.

[0089] Calibrate the actual high-latitude precipitation reanalysis data according to the expected high-latitude precipitation reanalysis data to obtain the calibrated high-latitude precipitation reanalysis data. This step improves the accuracy of the reanalysis data and reduces the uncertainty.

[0090] In the present invention, it is possible to ensure that the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data match the station-observed precipitation data in terms of spatial resolution, and through the establishment and application of a calibration model, the accuracy and consistency of these data are improved. Such processing is crucial for finally obtaining globally continuous precipitation data with a spatial distribution, because it allows data from different sources to be effectively stitched and synthesized on a global scale.

[0091] Optionally, the analysis of the station-observed precipitation data by the precipitation remote sensing calibration model to determine the expected low-latitude precipitation remote sensing data includes:

[0092] Obtain historical station-observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the grid cells of the station-observed precipitation data;

[0093] Calculate the first cumulative probability density value of the historical station-observed precipitation data of the grid cell, and obtain the historical low-latitude precipitation remote sensing data closest to the first cumulative probability density value among each historical low-latitude precipitation remote sensing data as the expected low-latitude precipitation remote sensing data.

[0094] In the present invention, it is necessary to obtain historical station-observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the grid cells of the station-observed precipitation data. These data are the basis for constructing the precipitation remote sensing calibration model.

[0095] Calculate the first cumulative probability density value of the historical station-observed precipitation data for each grid cell. The cumulative probability density value refers to the probability that a random variable is less than or equal to a certain specific value, which is an important concept in statistical analysis.

[0096] Obtain the historical low-latitude precipitation remote sensing data closest to the first cumulative probability density value calculated above among each historical low-latitude precipitation remote sensing data. This step is actually to find the data point in the historical remote sensing data that is most similar to the current station-observed data in terms of statistical characteristics as the expected low-latitude precipitation remote sensing data.

[0097] In the present invention, by using historical data and statistical methods to predict and calibrate the current low-latitude precipitation remote sensing data, the accuracy and reliability of the data are improved. This method combines the advantages of station-observed data and remote sensing data, and through the establishment and application of a statistical model, more accurate calibration and prediction of precipitation data are achieved.

[0098] Optionally, the analysis of the station-observed precipitation data by the precipitation reanalysis calibration model to determine the expected high-latitude precipitation reanalysis data includes:

[0099] Obtain historical station observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the grid cells of the station observed precipitation data;

[0100] Calculate the first cumulative probability density value of the historical station observed precipitation data of the grid cell, and obtain the historical high-latitude precipitation reanalysis data closest to the first cumulative probability density value in each historical high-latitude precipitation reanalysis data as the expected high-latitude precipitation reanalysis data.

[0101] In the present invention, historical station observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the grid cells of the station observed precipitation data are obtained. These historical data are the basis for constructing the precipitation reanalysis calibration model.

[0102] Calculate the first cumulative probability density value of the historical station observed precipitation data for each grid cell. The cumulative probability density value refers to the probability that a random variable is less than or equal to a certain specific value, which is an important concept in statistical analysis.

[0103] Obtain the historical high-latitude precipitation reanalysis data closest to the first cumulative probability density value calculated above in each historical high-latitude precipitation reanalysis data as the expected high-latitude precipitation reanalysis data. This step is actually to find the data point in the historical reanalysis data that is most similar to the current station observed data in statistical characteristics as the expected high-latitude precipitation reanalysis data.

[0104] In the present invention, historical data and statistical methods are used to predict and calibrate the current high-latitude precipitation reanalysis data, thereby improving the accuracy and reliability of the data. This method combines the advantages of station observed data and reanalysis data, and through the establishment and application of a statistical model, more accurate calibration and prediction of precipitation data are realized.

[0105] Optionally, splice and synthesize the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution, including:

[0106] Obtain the calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data for each grid cell; wherein, the grid cell is the spatial grid point corresponding to the gridded station observed precipitation data;

[0107] Based on the calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data of each grid cell, splice and synthesize to obtain globally continuous precipitation data in spatial distribution.

[0108] In the present invention, splicing and synthesis are performed based on the calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data of each grid cell. This process can be achieved by various methods, including but not limited to:

[0109] Stitch using the statistical characteristics of precipitation data, such as by methods like cumulative probability CDF matching.

[0110] Use machine learning algorithms, such as random forest, to merge multiple daily precipitation datasets, and adjust the merging results according to the interpretation rate obtained by the random forest to improve the accuracy.

[0111] Apply multi-source precipitation data fusion technology, which may include coupled merging and downscaling methods for station data, model data, and remote sensing data to improve the accuracy and resolution of the data.

[0112] In the present invention, precipitation datasets from different sources can be effectively stitched and synthesized to obtain a global precipitation dataset with high accuracy and high resolution, which is crucial for understanding and predicting the global water cycle, climate change, and related environmental problems.

[0113] Figure 2 Schematic diagram of the global precipitation data stitching and synthesis process provided for the present invention, as Figure 2 shown, includes:

[0114] Obtain global station-observed precipitation data, precipitation remote sensing data, and precipitation reanalysis data. Perform spatial interpolation on the globally station-observed precipitation data after quality control to obtain rasterized station-observed precipitation data,

[0115] whose spatial resolution is consistent with that of the precipitation remote sensing data and the precipitation reanalysis data (for example, the spatial resolution is 0.25°), effectively reducing the spatial sampling error of station observations. The spatial interpolation method can adopt the optimal interpolation method, thin plate spline interpolation method, etc. that can consider the influence of the precipitation spatial structure.

[0116] For precipitation remote sensing data in low-latitude regions, for each grid point, select an appropriate spatio-temporal window, and collect the matching ground observation data and precipitation remote sensing data. In practical applications, for example, a 10°×10° spatial range centered on the target grid point is used as the spatial window. The ground observation data is rasterized station-observed precipitation data, and the data of the observation stations in the spatio-temporal window are included in the statistics, that is, only the grid point precipitation values with at least one station observation are selected for statistics.

[0117] Establish a calibration model using the statistical relationship between the matching ground observation data and precipitation remote sensing data to complete the calibration of the precipitation remote sensing data.

[0118] The establishment of statistical relationships and calibration models can adopt methods such as linear regression, probability density matching, and machine learning. Among them, for the probability density matching method, for each precipitation remote sensing data grid point to be calibrated, the matching ground observation data and precipitation remote sensing data are collected in the spatio-temporal window, and their cumulative probability density distributions are respectively statistically analyzed. Calculate the cumulative probability density value of the precipitation remote sensing data grid point to be calibrated, and the corresponding ground observation data of this cumulative probability density value. The same precipitation cumulative probability density value corresponds to different ground observation data and precipitation remote sensing data, and the precipitation remote sensing data is calibrated according to the deviation between the two.

[0119] For the precipitation reanalysis data in high-latitude regions, for each of its grid points, an appropriate spatio-temporal window is selected to collect the matching ground observation data and precipitation reanalysis data. In practical applications, for example, a 10°×10° spatial range centered on the target grid point is used as the spatial window. The ground observation data is the rasterized precipitation data observed at stations, and the data of the observation stations in the spatio-temporal window are included in the statistics, that is, only the grid point precipitation values with at least one station observation are selected for statistics.

[0120] Use the statistical relationship between the matching ground observation data and precipitation reanalysis data to establish a calibration model to complete the calibration of the precipitation reanalysis data. The establishment of statistical relationships and calibration models can adopt methods such as linear regression, probability density matching, and machine learning. Among them,

[0121] For the probability density matching method, for each precipitation reanalysis data grid point to be calibrated, the matching ground observation data and precipitation reanalysis data are collected in the spatio-temporal window, and their cumulative probability density distributions are respectively statistically analyzed. Calculate the cumulative probability density value of the precipitation reanalysis data grid point to be calibrated, and the corresponding ground observation data of this cumulative probability density value. The same precipitation cumulative probability density value corresponds to different ground observation data and precipitation reanalysis data, and the precipitation reanalysis data is calibrated according to the deviation between the two.

[0122] Stitch and synthesize the calibrated precipitation remote sensing data in low-latitude regions and precipitation reanalysis data in high-latitude regions to obtain globally continuous precipitation data in terms of spatial distribution.

[0123] In the present invention, spatial interpolation and rasterization are performed on the precipitation data observed at stations, effectively reducing the spatial sampling error of station observations. For the precipitation remote sensing data in low-latitude regions and precipitation reanalysis data in high-latitude regions, calibration is performed through the rasterized precipitation data observed at stations, improving the spatio-temporal distribution consistency between precipitation remote sensing data and precipitation reanalysis data. Stitch and synthesize the calibrated precipitation remote sensing data in low-latitude regions and precipitation reanalysis data in high-latitude regions to obtain precipitation data with continuous spatial distribution and global coverage.

[0124] The global precipitation data stitching and synthesis device provided by the present invention will be described below. The global precipitation data stitching and synthesis device described below can be mutually corresponding and referred to the global precipitation data stitching and synthesis method described above.

[0125] Figure 3 FIG. is a schematic structural diagram of the global precipitation data stitching and synthesis device provided by the present invention, as Figure 3 shown, including:

[0126] The processing module 310 is used to perform spatial interpolation processing on the site ground observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data;

[0127] The first analysis module 320 is used to obtain low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the site observation precipitation data; wherein, the matched low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the site observation precipitation data;

[0128] The second analysis module 330 is used to calibrate the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site observation precipitation data according to the site observation precipitation data, to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0129] The synthesis module 340 is used to stitch and synthesize the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution.

[0130] In the present invention, by performing spatial interpolation processing on the ground observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data, the spatial sampling error of site observation can be effectively reduced, and the spatial resolution of site observation data can be made consistent with the spatial resolution of precipitation remote sensing data and precipitation reanalysis data. By obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each site observation precipitation data. This step ensures the spatial resolution consistency between different data sources and provides a basis for subsequent calibration and stitching. According to the rasterized site observation precipitation data, the matched low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are calibrated to obtain calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data. This step improves the data accuracy of precipitation remote sensing data and precipitation reanalysis data, as well as the consistency of their spatio-temporal distribution by establishing statistical relationships and calibration models. The calibrated low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data are stitched and synthesized to obtain globally continuous precipitation data in spatial distribution. The problem of inconsistent spatio-temporal distribution in the area near the stitching line of 50° (or 60°) north and south latitudes is solved, and global-scale precipitation analysis applications are realized.

[0131] Figure 4 is a schematic structural diagram of the electronic device provided by the present invention. As Figure 4 shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. Among them, the processor 410, the communications interface 420, and the memory 430 complete communication with each other through the communication bus 440. The processor 410 may call logic instructions in the memory 430 to execute the global precipitation data mosaicking and synthesis method, and the method includes: performing spatial interpolation processing on the site surface observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data;

[0132] obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the site observation precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the site observation precipitation data;

[0133] calibrating the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site observation precipitation data according to the site observation precipitation data to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0134] mosaicking and synthesizing the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution.

[0135] In addition, when the logic instructions in the above-mentioned memory 430 are implemented in the form of software function units and sold or used as an independent product, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0136] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the global precipitation data mosaicking and synthesizing method provided by each of the above methods. The method includes: performing spatial interpolation processing on the site ground observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data;

[0137] Obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the site observation precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the site observation precipitation data;

[0138] Calibrating the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site observation precipitation data according to the site observation precipitation data to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0139] Mosaicking and synthesizing the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution.

[0140] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the global precipitation data mosaicking and synthesizing method provided by each of the above methods. The method includes: performing spatial interpolation processing on the site ground observation precipitation data of global meteorological stations to obtain rasterized site observation precipitation data;

[0141] Obtaining low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match each of the site observation precipitation data; wherein, the matching low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the site observation precipitation data;

[0142] Calibrating the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site observation precipitation data according to the site observation precipitation data to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data;

[0143] Mosaicking and synthesizing the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain globally continuous precipitation data in spatial distribution.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0145] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for splicing and synthesizing global precipitation data, characterized in that: include: Perform spatial interpolation processing on the ground observation precipitation data of global meteorological stations to obtain rasterized station observation precipitation data; Acquire low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the observed precipitation data at each of the stations; wherein the matched low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the observed precipitation data at the stations; According to the precipitation data observed at the station, low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the precipitation data observed at the station are calibrated to obtain calibrated low-latitude precipitation remote sensing data and calibrated high-latitude precipitation reanalysis data; The calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data are spliced ​​and synthesized to obtain global precipitation data with continuous spatial distribution; According to the spatial resolution of the precipitation data observed at the site, the low-latitude precipitation remote sensing data and the high-latitude precipitation reanalysis data matching the spatial resolution of the precipitation data observed at the site are calibrated, including: The station observed precipitation data is analyzed by a precipitation remote sensing calibration model to determine expected low-latitude precipitation remote sensing data; wherein the precipitation remote sensing calibration model is constructed based on historical station observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the station observed precipitation data; Calibrate the low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain calibrated low-latitude precipitation remote sensing data; The station observed precipitation data are analyzed by a precipitation reanalysis calibration model to determine expected high-latitude precipitation reanalysis data; wherein the precipitation reanalysis calibration model is constructed based on historical station observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the station observed precipitation data; The high-latitude precipitation reanalysis data is calibrated according to the expected high-latitude precipitation reanalysis data to obtain calibrated high-latitude precipitation reanalysis data.

2. The global precipitation data splicing and synthesis method according to claim 1, characterized in that: The step of analyzing the observed precipitation data of the station by using the precipitation remote sensing calibration model to determine the expected low-latitude precipitation remote sensing data includes: Acquire historical site observation precipitation data and historical low-latitude precipitation remote sensing data corresponding to the grid points of the site observation precipitation data; The first cumulative probability density value of the historical site observation precipitation data of the grid grid points is calculated, and the historical low-latitude precipitation remote sensing data closest to the first cumulative probability density value in each historical low-latitude precipitation remote sensing data is obtained as the expected low-latitude precipitation remote sensing data.

3. The global precipitation data splicing and synthesis method according to claim 1, characterized in that: The step of analyzing the observed precipitation data of the station by using the precipitation reanalysis calibration model to determine the expected high-latitude precipitation reanalysis data includes: Acquire historical site observation precipitation data and historical high-latitude precipitation reanalysis data corresponding to the grid points of the site observation precipitation data; The first cumulative probability density value of the historical site observation precipitation data of the grid grid points is calculated, and the historical high-latitude precipitation reanalysis data closest to the first cumulative probability density value in each historical high-latitude precipitation reanalysis data is obtained as the expected high-latitude precipitation reanalysis data.

4. The global precipitation data splicing and synthesis method according to claim 1, characterized in that: The ground observation precipitation data of global meteorological stations are spatially interpolated to obtain rasterized station observation precipitation data, including: Performing data cleaning, outlier detection, and missing value filling on the site ground observation precipitation data of the global meteorological site to achieve quality control, and obtaining the site ground observation precipitation data after quality control; Based on a preset interpolation method, the quality-controlled site ground observation precipitation data is interpolated into a defined grid system to obtain rasterized site observation precipitation data; wherein the preset interpolation method includes at least one of the following: Kriging method, inverse distance weighting, and spline interpolation.

5. The global precipitation data splicing and synthesis method according to claim 1, characterized in that: The calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data are spliced ​​and synthesized to obtain global precipitation data with continuous spatial distribution, including: Obtaining calibrated low-latitude precipitation remote sensing data or calibrated high-latitude precipitation reanalysis data for each grid point; wherein the grid points are spatial grid points corresponding to the gridded site observation precipitation data; The calibrated low-latitude precipitation remote sensing data or the calibrated high-latitude precipitation reanalysis data of each grid point are spliced ​​and synthesized to obtain global precipitation data with continuous spatial distribution.

6. A global precipitation data splicing and synthesis device, characterized in that: include: A processing module is used to perform spatial interpolation processing on the ground observation precipitation data of global meteorological stations to obtain rasterized station observation precipitation data; The first analysis module is used to obtain low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the observed precipitation data of each of the stations; wherein the matched low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data have the same spatial resolution as the observed precipitation data of the station; The second analysis module is used to calibrate the low-latitude precipitation remote sensing data and high-latitude precipitation reanalysis data that match the spatial resolution of the site observation precipitation data according to the site observation precipitation data, so as to obtain the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data; A synthesis module, used for splicing and synthesizing the calibrated low-latitude precipitation remote sensing data and the calibrated high-latitude precipitation reanalysis data to obtain global precipitation data with continuous spatial distribution; Wherein, the device is also used for: The station observed precipitation data is analyzed by a precipitation remote sensing calibration model to determine expected low-latitude precipitation remote sensing data; wherein the precipitation remote sensing calibration model is constructed based on historical station observed precipitation data and historical low-latitude precipitation remote sensing data corresponding to the station observed precipitation data; Calibrate the low-latitude precipitation remote sensing data according to the expected low-latitude precipitation remote sensing data to obtain calibrated low-latitude precipitation remote sensing data; The station observed precipitation data are analyzed by a precipitation reanalysis calibration model to determine expected high-latitude precipitation reanalysis data; wherein the precipitation reanalysis calibration model is constructed based on historical station observed precipitation data and historical high-latitude precipitation reanalysis data corresponding to the station observed precipitation data; The high-latitude precipitation reanalysis data is calibrated according to the expected high-latitude precipitation reanalysis data to obtain calibrated high-latitude precipitation reanalysis data.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the global precipitation data splicing and synthesis method as claimed in any one of claims 1 to 5 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the global precipitation data splicing and synthesis method according to any one of claims 1 to 5 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the global precipitation data splicing and synthesis method according to any one of claims 1 to 5 is implemented.

Images