A method for temporal and spatial dynamic correction of satellite remote sensing precipitation products

By performing resolution matching and weight calculation on precipitation data from different data sources, the problem of insufficient precision and resolution of precipitation data in the prior art is solved, and high-precision and high-resolution precipitation data correction is achieved, which is suitable for multi-field applications .

CN112416915BActive Publication Date: 2024-12-03PEKING UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011257745.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-11-11
Publication Date
2024-12-03
Estimated Expiration
2040-11-11

AI Technical Summary

Technical Problem

The prior art is difficult to obtain high-precision and high-temporal and spatial resolution precipitation observation data that meets various needs, especially in complex terrain areas with low accuracy of ground rain meter data, while satellite data coverage and resolution are high but insufficient accuracy.

Method used

By acquiring data sets from different data sources (such as ground measured data and satellite remote sensing data), data matching with inconsistent resolutions is performed, the weight of low-precision data in high-precision data is calculated, and the data accuracy and resolution are improved. Specific methods include accumulation method, mobile window search method and proximity matching method, combining spatial and time weight calculations to obtain high-precision and high-resolution precipitation data.

Benefits of technology

It realizes more accurate precipitation data description, meets various needs of users, improves the accuracy and resolution of data, is suitable for precipitation data correction, and can be expanded to other fields such as traffic information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112416915B_ABST
    Figure CN112416915B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products. The method comprises: obtaining a first data set and a second data set from different data sources for describing the same object; matching a high-resolution data set with a low-resolution data set based on data observation dimensions with inconsistent resolutions; based on the matched data sets, calculating the weight of each data in the low-precision data set in the corresponding matched high-precision data from different data observation dimensions; calculating based on the data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution. The method of the present invention provides users with the required high-accuracy data and improves user experience. At the same time, the method provided by the present invention is low-cost and simple and easy to implement.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data correction. More specifically, the embodiments of the present invention relate to a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products. Background Art

[0002] This section is intended to provide a background or context to embodiments of the invention that are recited in the claims. No description herein is admitted to be prior art by inclusion in this section.

[0003] Currently, users can obtain different observation data of an object through different observation devices. However, since these observation data come from different data sources, the multiple data sets obtained by users usually have different characteristics, such as higher data accuracy or higher resolution. At the same time, since the obtained data will be affected to varying degrees by various factors, the obtained data sets often cannot meet the multiple needs of users at the same time.

[0004] For example, currently, when users want to obtain regional precipitation data, the most direct way is to obtain precipitation data through ground rain gauges. However, due to the influence of factors such as terrain and climate, it is difficult to accurately describe large regions, oceans, mountains and other complex terrain areas. If precipitation data is obtained through the Global Precipitation Program observation satellite, although precipitation data with wider coverage and higher resolution can be obtained, the regional observation accuracy is relatively low and the error is large.

[0005] Therefore, obtaining data that can meet the various needs of users is the development direction of this field. At the same time, how to obtain high-precision, high-temporal and high-spatial resolution precipitation observation data is also an urgent problem to be solved in the current precipitation research field. Summary of the invention

[0006] A brief overview of the disclosure is given below in order to provide a basic understanding of certain aspects of the disclosure. It should be understood that this overview is not an exhaustive overview of the disclosure. It is not intended to identify key or important parts of the disclosure, nor is it intended to limit the scope of the disclosure. Its purpose is simply to give certain concepts in a simplified form as a prelude to a more detailed description discussed later.

[0007] In view of the above-mentioned problems in the prior art, one of the purposes of the present disclosure is to provide a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products, so as to at least solve the problems existing in the current prior art.

[0008] In a first aspect of an embodiment of the present invention, a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products is provided, comprising:

[0009] Acquire a first data set and a second data set from different data sources for describing the same object, wherein data accuracy of the first data set is higher than that of the second data set, and resolution of at least one data observation dimension of the first data set is different from that of the second data set;

[0010] Matching high-resolution datasets with low-resolution datasets based on the observation dimensions of data with inconsistent resolutions;

[0011] Based on the matched data set, the weight of each data in the low-precision data set in the corresponding matched high-precision data is calculated from different data observation dimensions;

[0012] Calculation is performed based on a data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution.

[0013] In a second aspect of the embodiments of the present invention, a method for spatiotemporal dynamic correction of precipitation satellite remote sensing precipitation products is provided, which uses any one of the methods described in the first aspect to correct precipitation data based on ground measured precipitation data and satellite remote sensing precipitation data.

[0014] In an example of this implementation, the ground measured precipitation data is an APH (APHRODITE) dataset, and the satellite remote sensing precipitation data is an IMERG (GPM-IMERG) dataset.

[0015] In an example of this implementation manner, the data accuracy of the APH dataset is higher than that of the IMERG dataset, and the temporal and spatial resolution of the APH dataset is lower than that of the IMERG dataset.

[0016] According to the spatiotemporal dynamic correction method of satellite remote sensing precipitation products in the embodiment of the present invention, two data sets of different precision and resolution from different data sources are matched, and based on this, the weight of each data in the low-precision data set in the corresponding matched high-precision data is calculated from different data observation dimensions, and then the data set with high data precision and the weight calculated from different data observation dimensions are calculated to obtain data with high data precision and high data observation resolution, thereby improving the precision of the data required by the user, providing the user with more accurate data support, and improving the user experience. At the same time, the method provided by the present invention is low-cost and simple and easy to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0018] Figure 1 A schematic diagram of an application scenario of a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products provided in one embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a flow chart of a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products provided in one embodiment of the present invention;

[0020] Figure 3 The Taylor diagram numerical distribution diagram of the six validation sub-areas before and after data correction;

[0021] Figure 4a-4f for Figure 3 The numerical time series changes of precipitation products in each sub-region before and after data correction are shown;

[0022] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0023] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0024] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0025] According to an embodiment of the present invention, a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products is proposed.

[0026] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0027] The principle and spirit of the present invention are explained in detail below with reference to several representative embodiments of the present invention.

[0028] Application Scenario Overview

[0029] First reference Figure 1In the figure, multiple users can provide multiple data sets with different accuracy and resolution from different data sources for the device that can execute the spatiotemporal dynamic correction method for satellite remote sensing precipitation products, or the device that can execute the spatiotemporal dynamic correction method for satellite remote sensing precipitation products can also obtain the data sets required by users through other means such as the Internet. This is not limited here, as long as the resolution and accuracy of the data sets obtained by the device are different.

[0030] The device that can execute the spatiotemporal dynamic correction method of satellite remote sensing precipitation products will match the acquired data sets and calculate weights from different observation dimensions, and based on the high-precision data sets in the acquired data sets and the weights of different observation dimensions, provide users with the required high-precision and high-resolution data through calculation.

[0031] Exemplary Methods

[0032] Combine the following Figure 1 For application scenarios, refer to Figure 2 The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to an exemplary embodiment of the present invention is described. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present invention, and the embodiments of the present invention are not limited in this regard. On the contrary, the embodiments of the present invention can be applied to any applicable scenario.

[0033] The present invention is further described below in conjunction with specific implementation cases.

[0034] An embodiment of the present invention provides a method for spatiotemporal dynamic correction of satellite remote sensing precipitation products, the method comprising:

[0035] Step S210, obtaining a first data set and a second data set from different data sources for describing the same object, wherein the data accuracy of the first data set is higher than the data accuracy of the second data set, and the resolution of at least one data observation dimension of the first data set is different from that of the second data set;

[0036] Step S220, matching the data set with a high resolution with the data set with a low resolution based on the data observation dimensions with inconsistent resolutions;

[0037] Step S230, based on the matched data set, the weight of each data in the low-precision data set in the corresponding matched high-precision data is calculated from different data observation dimensions;

[0038] Step S240, performing calculation based on the data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution.

[0039] The data obtained by the method of this embodiment is mainly used for the spatiotemporal dynamic correction of satellite remote sensing precipitation products, which can provide users with high data accuracy and high data observation resolution observation data for a certain observation object, thereby meeting the comprehensive needs of users. At the same time, it also makes up for the serious deficiencies of the observation data caused by the data source itself, and obtains a more accurate description.

[0040] This implementation is not only applicable to precipitation data correction, but also can be applied to the fields of transportation, medical treatment, etc. The spatiotemporal dynamic correction method of satellite remote sensing precipitation products provided in this application provides users with more accurate data by correcting data from different data sources.

[0041] The following is an explanation of how to perform spatiotemporal dynamic correction of satellite remote sensing precipitation products with the help of attached figures:

[0042] First, in step S210, a first data set and a second data set from different data sources for describing the same object are obtained. The acquisition method can be configured to be acquired through user input, or can also be configured to be automatically acquired through the Internet or other means. The data accuracy of the first data set is higher than the data accuracy of the second data set, and the resolution of at least one data observation dimension of the first data set is different from that of the second data set. The data set can be one of data describing road traffic information, the total number of white blood cells contained in the blood, or regional precipitation data. In the following embodiments of the present application, precipitation data is taken as an example for detailed description.

[0043] Next, step S220 is performed to match the high-resolution data set with the low-resolution data set based on the data observation dimensions with inconsistent resolutions. Since the observation dimensions of the data sets obtained by different data observation methods, the number of values ​​contained, and the user's requirements for the result data are different, the corresponding data matching method is selected according to the observation dimensions and / or values ​​of the data. The data matching method includes an accumulation method (applicable to data matching with inconsistent resolutions in the time observation dimension), a moving window search method, and a proximity matching method (applicable to data matching with inconsistent resolutions in the space observation dimension). When matching the data sets, first, the high (time) resolution data is accumulated so that its resolution is consistent with the corresponding low (time) resolution data set to be matched. For example, the time resolution of data set A is hours, that is, the minimum description unit of each data in data set A in the time dimension is hours, and the time resolution of data set B is minutes, that is, the minimum description unit of each data in data set B in the time dimension is minutes. Obviously, the time resolution accuracy of data set B is higher than that of data set A. When performing accumulation matching, the data of each minute in a certain hour in data set B is accumulated, so that the time resolution of the data in data set B is consistent with that of data set A.

[0044] In another embodiment of the present embodiment, data matching is further performed by a moving window method. Specifically, each data in the high-resolution data set is used as the window center, and the high-resolution data currently located at the window center is matched one by one with the nearest low-resolution data through a moving window. The size of the moving window can be preset by the user, or a commonly used size in the field is used as the preset size; again, the area of ​​each low-resolution data covered by the moving window and the proportion of the area of ​​each low-resolution data falling into the window to the total area of ​​the low-resolution data are calculated; finally, after traversing each data of all high-resolution data, the low-resolution data with the largest proportion obtained and the current high-resolution data at the center of the moving window are used as matching data. When the value of the current data at the center of the moving window is a null value, the proximity matching method is used for data matching.

[0045] In the above embodiment, the high resolution in the high-resolution data set generally refers to the spatial resolution, so the size of the high resolution is smaller than the size of the low-resolution data. When setting the size of the window, it can be set to include several high-resolution data, which are smaller than the size of a single low-resolution data. For specific examples, please refer to the precipitation data correction example below.

[0046] Next, step S230 is executed. Based on the matched data set, the weight of each data in the low-precision data set in the corresponding high-precision data is calculated from different data observation dimensions; when the data set is matched using the accumulation method, the ratio of the value of each low-precision data to the value of the accumulated data to which it belongs is calculated, and the ratio is used as the first weight of the low-precision data in the high-precision data to which it corresponds. When matching data is obtained based on the moving window search method and the proximity matching method, the ratio of the value of each low-precision data to the sum of the values ​​of multiple low-precision data that are the same as the high-precision data to which it corresponds is calculated, and the ratio is used as the second weight of the low-precision data in the high-precision data to which it corresponds. Taking into account that data with a value of zero will inevitably appear in the data set, in order to ensure the accuracy of the data finally corrected, the first data set and the second data set can be processed with a value of 0 based on the matched data, that is, when the value of the low-resolution data matching the high-resolution data is 0, and the high-resolution data is not 0, the value of the high-resolution data at this time is evenly distributed to the multiple low-precision data that match it; when the low-resolution data and the high-resolution data are both 0, the value of the high-resolution data is assigned to the low-precision data, so the value of the low-resolution data is 0 at this time.

[0047] Finally, step S240 is performed to perform a multiplication calculation based on the data set with high data precision and the weights calculated from different data observation dimensions to obtain the data with high data precision and high data observation resolution required by the user.

[0048] Specifically, the data correction method can be applied to different fields, and is described by taking the correction of precipitation observation data as an example.

[0049] For example, China (73°E-136°E, 3°N-54°N) was selected as the research area, and the IMERG high-temporal-resolution precipitation dataset and the APHRODITE high-precision ground-based daily precipitation dataset from 2007 to 2015 were subjected to spatiotemporal correction, and finally a high-quality high-temporal-resolution precipitation product with a temporal resolution of 30 minutes and a spatial resolution of 0.1°×0.1° was obtained. The resolution of the IMERG data is different from that of the APHRODITE data, and the accuracy of the APHRODITE data is higher than that of the IMERG data.

[0050] As in step S210, obtain the GPM satellite IMERG remote sensing precipitation data and APHRODITE high-precision daily precipitation data of the study area; the GPM IMERG data is the Final-Run V06 version of the global remote sensing precipitation product, which can be downloaded from the official website of the National Aeronautics and Space Administration (NASA) Precipitation Measurement Mission (https: / / pmm.nasa.gov / precipitation-measurement-missions) from June 2000 to the present. The L3 precipitation data can also be configured to download and update the required GPM IMERG data through the website. This product is calibrated using monthly-scale data from ground stations. Its accuracy is closer to the actual precipitation data and it is a scientific research-level product with a spatial resolution of 0.1°×0.1° and a temporal resolution of 30 min. The APHRODITE data is a V1101EX_R1 high-resolution daily precipitation product (mm / day) with a spatial resolution of 0.25°×0.25° and a temporal resolution of 1 day. The precipitation data from 2007 to 2015 can be obtained in the same way on the APHRODITE official website (http: / / aphrodite.st.hirosaki-u.ac.jp / download / ). Both data formats are in NetCDF format.

[0051] According to step S220, the above data is preprocessed: the NetCDF data is read using the corresponding tool in the Matlab software and processed into a raster format output; because the time resolution of the IMERG product is 30 minutes, a total of 48 IMERG raster layers can be obtained every day, and the non-empty raster pixel values ​​of all IMERG products on that day are accumulated as the pixel values ​​corresponding to the daily resolution, and the time resolution of the IMERG precipitation data obtained in step S210 is processed as daily; the APH pixel closest to the current IMERG pixel is searched through a moving window calculation and matched one by one. The specific steps of the moving window method for searching and matching are as follows: each pixel of the IMERG data product is taken as the window center, and the size of the window is It is set to 0.3°*0.3°, that is, a moving window contains 9 IMERG pixels. For each APH pixel covered by the moving window, the proportion of the area falling into the moving window to its total area is calculated, and the APH pixel with the largest area ratio is taken as the nearest pixel of the central IMERG pixel of the moving window and matched; the null value pixels in the APHRODITE pixel data are obtained by conditional screening, and the neighboring matching processing is performed on them. This is for the case where the APH pixels near the land-sea boundary and in some areas are null values, and the adjacent pixel grids are selected nearby to assign values ​​to the null value pixels; finally, the daily IMERG precipitation data and APHRODITE precipitation data after all pixels are matched are obtained.

[0052] According to step S230, the paired data processed in step S220 is used as a benchmark. When the raster pixel values ​​of the IMERG data and the APH data are not 0, that is, both indicate precipitation, the spatial weight of a single IMERG pixel in the corresponding APH pixel is calculated using a specific spatial weight calculation method and constrained by specific conditions. At the same time, a specific time weight calculation method is used to assign a time weight corresponding to each 30-minute pixel data within a day.

[0053] The weight calculation method for a specific space is in the form of Wsi=Si / (S1+…+Sn);

[0054] Where Wsi represents the spatial weight of the i-th IMERG pixel; Si represents the daily precipitation of the i-th IMERG pixel; S1+…+Sn represents the sum of the daily precipitation of n IMERG pixels belonging to a specific APH pixel;

[0055] The weight calculation method for a specific time is in the form of Wti=Ti / (T1+…+T48);

[0056] Where Wti represents the time weight of the ith IMERG pixel; Ti represents the precipitation of the ith half-hour IMERG pixel; and T1+…+T48 represents the sum of the precipitation of the 48 half-hour IMERG pixels belonging to a specific IMERG pixel on that day.

[0057] For other situations, corresponding rules are formulated and value assignment calculations are performed to establish a data processing model. The specific rules are as follows: when the APH grid pixel value is not 0, which means there is precipitation, and the daily resolution IMERG grid pixel value is 0, which means there is no precipitation, the daily precipitation of the APH pixel is evenly distributed to the 48 matched 30-min resolution IMERG pixels; when the APH pixel data indicates there is no precipitation, all matched 30-min resolution IMERG pixels are assigned a value of 0.

[0058] Finally, the obtained data is corrected according to step S240: based on step S230, the corresponding spatiotemporal weights when both IMERG data and APH data indicate precipitation at a spatial scale of 0.1°×0.1° and a time scale of 30 minutes are determined, and then the IMERG satellite remote sensing precipitation product is corrected based on the daily APH precipitation product at a scale of 0.25°×0.25°, and the precipitation fusion data with a time resolution of 30 minutes and a spatial resolution of 0.1°×0.1° after total amount control is calculated. The calculation method is Wi=A·Wsi·Wti

[0059] Among them, Wi represents the corrected precipitation data of the i-th pixel, with a temporal resolution of 30 minutes and a spatial resolution of 0.1°×0.1°; A represents the precipitation data of the specific APH pixel with a temporal resolution of 1 day and a spatial resolution of 0.25°×0.25°; Wsi represents the spatial weight of the i-th IMERG pixel; Wti represents the temporal weight of the i-th IMERG pixel.

[0060] Finally, a high-precision, high-temporal-resolution precipitation product raster dataset for some regions of China from 2007 to 2015 was obtained. The target area was used as the validation area and divided into 6 sub-areas. The validation results show that after the data correction operation based on the STWMRC algorithm (i.e., the method proposed in this application) was completed, the precipitation products of each region were closer to the precipitation information of the ground station in each sub-area (e.g. Figure 3 The numerical deviation is significantly reduced (as shown in Figure 4a-4f This fully demonstrates the availability and reliability of the corrected precipitation raster data products with high precision and high temporal and spatial resolution.

[0061] The data correction method of the present invention can also be used in the field of traffic information to correct the traffic data of automobiles, ships, aircraft, etc., and provide users with accurate traffic information data. Specifically, users can obtain the required observation data sets A and B of vehicle traffic from different data sources through the Internet and other means; the data accuracy (spatial data) of vehicle popular information provided by the data sets A and B is different and the resolution (time data) of at least one data observation dimension is also different; match the data sets A and B with inconsistent resolutions; then, based on the matched data sets, calculate the weight of each data in the low-precision data set in the corresponding matched high-precision data for different data observation dimensions; finally, calculate and obtain the high-precision, high-resolution data required by the user.

[0062] In the above embodiment, the method of the present invention matches data set A and data set B by the accumulation method, the moving window search method and the proximity matching method, and performs 0 value processing on data set A and data set B. The specific method is not repeated here. The paired data is used as a benchmark to calculate the spatial weight and the time weight. The calculation formula of the spatial weight is:

[0063] Wsi=Si / (S1+…+Sn)

[0064] Among them, Wsi represents the spatial weight of the i-th low-precision dataset pixel; Si represents the daily traffic flow of the i-th low-precision dataset pixel; S1+...+Sn represents the sum of the daily traffic flow of n low-precision dataset pixels belonging to a specific high-precision dataset pixel.

[0065] The calculation formula of time weight is:

[0066] Wti=Ti / (T1+…+T48);

[0067] Among them, Wti represents the time weight of the i-th low-precision dataset pixel; Ti represents the traffic volume of the i-th half-hour low-precision dataset pixel; T1+…+T48 represents the sum of the 48 half-hour IMERG pixel traffic volumes of the low-precision dataset pixel on that day belonging to a specific high-precision dataset.

[0068] Finally, the correction data required by the user is obtained by calculation based on the high-precision data and the corresponding spatial weight and temporal weight. The calculation formula is:

[0069] Wi=A·Wsi·Wti

[0070] Among them, A is the high-precision data in the acquired traffic flow data set.

[0071] In general, the present application provides users with the required high-precision and high-resolution data by performing correction calculations on two data sets of the same object obtained from different data sources, thereby meeting user needs and improving user experience. At the same time, the method provided by the present invention is low-cost and simple to implement, saving the cost of users obtaining the required data.

[0072] In the description of the present invention, it should be noted that the terms “first”, “second” and “third” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0073] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0074] In addition, although the operations of the method of the present invention are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0075] In summary, in the embodiments according to the present invention, the present invention provides the following technical solutions:

[0076] 1. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products, comprising:

[0077] Acquire a first data set and a second data set from different data sources for describing the same object, wherein data accuracy of the first data set is higher than that of the second data set, and resolution of at least one data observation dimension of the first data set is different from that of the second data set;

[0078] Matching high-resolution datasets with low-resolution datasets based on the observation dimensions of data with inconsistent resolutions;

[0079] Based on the matched data set, the weight of each data in the low-precision data set in the corresponding matched high-precision data is calculated from different data observation dimensions;

[0080] Calculation is performed based on a data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution.

[0081] 2. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 1, wherein the high-resolution data set is matched with the low-resolution data set based on the data observation dimensions with inconsistent resolutions, including:

[0082] According to the observed dimension and / or value of the data, the corresponding data matching method is selected to match the high-resolution dataset with the low-resolution dataset.

[0083] 3. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Option 2, wherein the data matching method includes an accumulation method, a moving window search method, and a proximity matching method.

[0084] 4. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 3, wherein the data matching using the accumulation method includes:

[0085] Determine the amount of high-resolution data included in a corresponding higher-resolution data set based on the resolution of a lower-resolution data set of a certain data observation dimension;

[0086] A corresponding amount of high-resolution data is accumulated to form low-resolution data that is consistent with the resolution of the data set with a lower resolution of the observation dimension of the initial data.

[0087] 5. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in solution 3 or 4, wherein the data matching using a moving window search method comprises:

[0088] Taking each data of the high-resolution data as the window center, a moving window of a preset size is used to search and match the low-resolution data closest to the high-resolution data currently located at the center of the window one by one;

[0089] For each low-resolution data covered by the moving window, calculate the ratio of the area of ​​the current low-resolution data falling within the window to the total area of ​​the low-resolution data;

[0090] The low-resolution data with the largest ratio and the current high-resolution data at the center of the moving window are used as matching data.

[0091] 6. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in any one of Schemes 3-5, wherein when data matching is performed using a moving window search method, if the value of the current data at the center of the window is empty, a proximity matching method is used for data matching.

[0092] 7. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 4, wherein the weight of the matching data obtained by the accumulation method is calculated based on the following method:

[0093] Calculate the ratio of the value of each low-precision data to the value of the accumulated data to which it belongs;

[0094] The ratio is used as the first weight of the low-precision data in its corresponding high-precision data.

[0095] 8. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 5 or 6, wherein the weights of the matching data obtained by the moving window search method and the proximity matching method are calculated based on the following method:

[0096] Calculate the ratio of the value of each low-precision data to the sum of the values ​​of multiple low-precision data that are the same as the corresponding high-precision data;

[0097] The ratio is used as the second weight of the low-precision data in its corresponding high-precision data.

[0098] 9. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to any one of schemes 1 to 8, wherein before calculating based on a data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution, the method further comprises:

[0099] The data with a value of 0 in the first data set and the second data set are processed respectively based on the matched data.

[0100] 10. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Solution 9, wherein when the value of the low-precision data is 0, the low-precision data is assigned a value based on its matching high-precision data;

[0101] When the value of the high-precision data is 0, the value of the low-precision data that matches it is assigned 0.

[0102] 11. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to solution 10, wherein when the value of the low-precision data is 0, assigning a value to the low-precision data based on its matching high-precision data comprises:

[0103] Evenly distribute the value of high-precision data to the multiple low-precision data that it matches.

[0104] 12. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to any one of schemes 1 to 11, wherein the calculation is performed based on a data set with high data accuracy and the weights calculated from different data observation dimensions to obtain data with high data accuracy and high data observation resolution, including:

[0105] A multiplication calculation is performed based on the high-precision data and the corresponding first weight and second weight to obtain data with high data precision and high data observation resolution.

[0106] 13. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products, wherein precipitation data correction is performed based on ground measured precipitation data and satellite remote sensing precipitation data using any of the methods described in Schemes 1-12.

[0107] 14. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 13, wherein the ground measured precipitation data is an APH (APHRODITE) dataset, and the satellite remote sensing precipitation data is an IMERG (GPM-IMERG) dataset.

[0108] 15. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 14, wherein the data accuracy of the APH dataset is higher than that of the IMERG dataset, and the spatiotemporal resolution of the APH dataset is lower than that of the IMERG dataset.

[0109] 16. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 15, wherein the spatiotemporal resolution of the APH dataset is 1 day and 0.25°, and the spatiotemporal resolution of the IMERG dataset is 30 minutes and 0.1°.

[0110] 17. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in any one of Schemes 14-16, wherein the temporal resolution of the IMERG dataset is processed to 1 day by an accumulation method so that the APH dataset and the IMERG dataset match in temporal resolution;

[0111] The APH dataset and IMERG dataset were matched in spatial resolution using a moving window search method.

[0112] 18. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 17, wherein the weight of the spatial observation dimension is calculated by:

[0113] Wsi=Si / (S1+…+Sn);

[0114] Where Wsi represents the spatial weight of the i-th IMERG pixel; Si represents the daily precipitation of the i-th IMERG pixel; S1+…+Sn represents the sum of the daily precipitation of n IMERG pixels matching a specific APH pixel;

[0115] The weight of the time observation dimension is calculated as follows:

[0116] Wti=Ti / (T1+…+T48);

[0117] Where Wti represents the time weight of the i-th IMERG pixel; Ti represents the precipitation of the i-th 30-minute IMERG pixel; and T1+…+T48 represents the sum of the precipitation of the 48 30-minute IMERG pixels belonging to the IMERG pixel on a specific day.

[0118] 19. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 17 or 18, wherein when the APH pixel data indicates precipitation but the matched IMERG pixel does not indicate precipitation, the daily precipitation of the APH pixel is evenly distributed among the matched IMERG pixels; and when the APH pixel data indicates no precipitation, the matched IMERG pixel is assigned a value of 0.

[0119] 20. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as described in Scheme 19, wherein the calculation method for data correction is:

[0120] Wi = A·Wsi·Wti;

[0121] Among them, Wi represents the corrected precipitation data of the i-th pixel, with a temporal resolution of 30 minutes and a spatial resolution of 0.1°×0.1°; A represents the precipitation data of a specific APH pixel with a temporal resolution of 1 day and a spatial resolution of 0.25°×0.25°; Wsi represents the spatial weight of the i-th IMERG pixel; Wti represents the temporal weight of the i-th IMERG pixel.

Claims

1. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products. include: Acquire a first data set and a second data set from different data sources for describing the same object, wherein data accuracy of the first data set is higher than that of the second data set, and a resolution of at least one data observation dimension of the first data set is different from that of the second data set; Matching high-resolution datasets with low-resolution datasets based on the observation dimensions of data with inconsistent resolutions; Based on the matched data set, the weight of each data in the low-precision data set in the corresponding matched high-precision data is calculated from different data observation dimensions; Calculating based on a data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution; The matching of a high-resolution data set with a low-resolution data set based on data observation dimensions with inconsistent resolutions includes: According to the observation dimension and / or value of the data, a corresponding data matching method is selected to match the data set with high resolution with the data set with low resolution; the data matching method includes an accumulation method, a moving window search method and a proximity matching method; Among them, the moving window search method is used for data matching, including: Taking each data of the high-resolution data as the window center, a moving window of a preset size is used to search and match the low-resolution data closest to the high-resolution data currently located at the center of the window one by one; For each low-resolution data covered by the moving window, calculate the ratio of the area of ​​the current low-resolution data falling within the window to the total area of ​​the low-resolution data; The low-resolution data with the largest ratio and the current high-resolution data at the center of the moving window are used as matching data; Among them, when data matching is performed by the moving window search method, if the value of the current data at the center of the window is empty, the proximity matching method is used for data matching.

2. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to claim 1, in, Data matching using the accumulation method includes: Determine the amount of high-resolution data included in a corresponding higher-resolution data set based on the resolution of a lower-resolution data set of a certain data observation dimension; A corresponding amount of high-resolution data is accumulated to form low-resolution data that is consistent with the resolution of the data set with a lower resolution of the observation dimension of the initial data.

3. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as claimed in claim 2, in, The weight of the matching data obtained by the cumulative matching method is calculated based on the following method: Calculate the ratio of the value of each low-precision data to the value of the accumulated data to which it belongs; The ratio is used as the first weight of the low-precision data in its corresponding high-precision data.

4. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to claim 1, in, The weights of the matching data obtained by the moving window search method and the proximity matching method are calculated based on the following method: Calculate the ratio of the value of each low-precision data to the sum of the values ​​of multiple low-precision data that are the same as the corresponding high-precision data; The ratio is used as the second weight of the low-precision data in its corresponding high-precision data.

5. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to any one of claims 1 to 4, in, Before calculating based on the data set with high data precision and the weights calculated from different data observation dimensions to obtain data with high data precision and high data observation resolution, the method further includes: The data with a value of 0 in the first data set and the second data set are processed respectively based on the matched data.

6. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to claim 5, in, When the value of the low-precision data is 0, assigning a value to the low-precision data based on its matching high-precision data; When the value of the high-precision data is 0, the value of the low-precision data that matches it is assigned 0.

7. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products according to claim 6, in, When the value of the low-precision data is 0, assigning a value to the low-precision data based on the high-precision data that matches it includes: Evenly distribute the value of high-precision data to the multiple low-precision data that it matches.

8. The method for spatiotemporal dynamic correction of satellite remote sensing precipitation products as claimed in claim 7, in, The data set with high data precision and the weights calculated from different data observation dimensions are used for calculation to obtain data with high data precision and high data observation resolution, including: A multiplication calculation is performed based on the high-precision data and the corresponding first weight and second weight to obtain data with high data precision and high data observation resolution.

9. A method for spatiotemporal dynamic correction of satellite remote sensing precipitation products. in, The method described in any one of claims 1 to 8 is used to correct precipitation data based on ground-measured precipitation data and satellite remote sensing precipitation data; wherein the ground-measured precipitation data is an APH (APHRODITE) dataset, and the satellite remote sensing precipitation data is an IMERG (GPM-IMERG) dataset; the data accuracy of the APH dataset is higher than that of the IMERG dataset, and the temporal and spatial resolution of the APH dataset is lower than that of the IMERG dataset; the temporal and spatial resolutions of the APH dataset are 1 day and 0.25°, respectively, and the temporal and spatial resolutions of the IMERG dataset are 30 minutes and 0.1°, respectively; wherein the temporal resolution of the IMERG dataset is processed to 1 day by an accumulation method, so that the APH dataset and the IMERG dataset match in temporal resolution; The APH dataset and the IMERG dataset are matched in spatial resolution by a moving window search method; the weight of the spatial observation dimension is calculated as follows: Wsi=Si / (S1+…+Sn); Where Wsi represents the spatial weight of the i-th IMERG pixel; Si represents the daily precipitation of the i-th IMERG pixel; S1+…+Sn represents the sum of the daily precipitation of n IMERG pixels matching a specific APH pixel; The weight of the time observation dimension is calculated as follows: Wti=Ti / (T1+…+T48); Where Wti represents the time weight of the i-th IMERG pixel; Ti represents the precipitation of the i-th 30-minute IMERG pixel; T1+…+T48 represents the sum of the precipitation of the 48 30-minute IMERG pixels belonging to the IMERG pixel on a specific day; When the APH pixel data has precipitation and the matched IMERG pixel has no precipitation, the daily precipitation of the APH pixel is evenly distributed to the matched IMERG pixel; when the APH pixel data has no precipitation, the matched IMERG pixel is assigned 0; The calculation method of data correction is: Wi = A·Wsi·Wti; Among them, Wi represents the corrected precipitation data of the i-th pixel, with a temporal resolution of 30 minutes and a spatial resolution of 0.1°×0.1°; A represents the precipitation data of a specific APH pixel with a temporal resolution of 1 day and a spatial resolution of 0.25°×0.25°; Wsi represents the spatial weight of the i-th IMERG pixel; Wti represents the temporal weight of the i-th IMERG pixel.