Multi-source rainfall information time-varying fusion method, device and system considering error variation

Abstract

The invention discloses a multi-source rainfall information time-varying fusion method, device and system considering error variation. The method comprises the steps of generating a first satellite inversion rainfall product, a second satellite inversion rainfall product and a reanalysis simulation rainfall product which have the same temporal-spatial resolution; at each time point of each rainfall station, determining time-varying optimal weights of the first satellite inversion rainfall product, the second satellite inversion rainfall product and the reanalysis simulation rainfall product by taking an error between a minimized fusion rainfall product and rainfall station observation data as a target; performing interpolation processing on the time-varying optimal weight by using an inverse distance weighting method to generate a time-varying fusion weight; and generating a fused precipitation product based on the first satellite inversion precipitation product, the second satellite inversion precipitation product, the reanalysis simulation precipitation product and the time-varying fusion weight. According to the method, time-varying fusion of multi-source rainfall information can be effectively realized, and data support is provided for data-free drainage basin water resource management and water disaster monitoring and early warning.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-source precipitation information fusion, and specifically relates to a multi-source precipitation information time-varying fusion method, device and system taking error variation into consideration. Background Art

[0002] High-quality precipitation data is an important basic data for global water cycle simulation, regional climate monitoring, watershed water disaster forecasting and warning, and water conservancy project planning and design. Traditional precipitation data mainly relies on ground rain gauge observations, but the rain gauge network is usually sparsely distributed and uneven, making it difficult to reliably characterize the temporal and spatial variation characteristics of actual precipitation, and cannot meet the requirements of high-precision hydrological simulation for engineering applications.

[0003] With the rapid development of satellite remote sensing technology and data inversion algorithms, precipitation products based on satellite remote sensing quantitative estimation have the advantages of wide coverage, strong spatial continuity, and high spatiotemporal resolution. They effectively overcome the spatial limitations of ground rain gauges and solve the problem of precipitation observation in areas without rain gauges, providing potential precipitation data references for hydrological model simulations and drought and flood monitoring and early warning in areas without data. At the same time, based on observation data from ground rain gauges, meteorological radars, drones, and satellites, scholars have used numerical models of land and climate processes, analysis and prediction systems, and data assimilation technology to generate spatiotemporal continuous reanalysis products, which can provide global coverage, high spatiotemporal resolution, and long time span precipitation data.

[0004] However, satellite precipitation products and reanalysis precipitation products have large uncertainties, especially in areas with complex terrain and climate, due to factors such as observation, cloud characteristics, inversion algorithms, assimilation schemes, and model parameterization processes. Most scholars have effectively improved the reliability of precipitation estimates by fusing multi-source data. However, existing studies generally ignore the true variability of precipitation on small time scales, rarely consider the time-varying characteristics of fusion weights, and fail to efficiently integrate the advantages of various precipitation information, which restricts basin water resources planning and management and the joint operation and scheduling of reservoir groups. Summary of the invention

[0005] In response to the above problems, the present invention proposes a time-varying fusion method, device and system for multi-source precipitation information taking into account error variation, which can effectively fuse multi-source precipitation information in a time-varying manner and provide more accurate and highly operational data support for water resources management and water disaster monitoring and early warning in data-free river basins.

[0006] In order to achieve the above technical objectives and the above technical effects, the present invention is implemented through the following technical solutions:

[0007] In a first aspect, the present invention provides a method for time-varying fusion of multi-source precipitation information considering error variation, comprising:

[0008] Generate the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product with the same temporal and spatial resolution;

[0009] At each time point of each rain gauge station, the time-varying optimal weights of the first satellite-inverted precipitation product, the second satellite-inverted precipitation product and the reanalysis simulated precipitation product are determined with the goal of minimizing the error between the fused precipitation product and the rain gauge station observation data;

[0010] Using an inverse distance weighted method to interpolate the time-varying optimal weights to generate time-varying fusion weights, wherein the time-varying fusion weights have the same temporal and spatial resolution as the three precipitation products;

[0011] A fused precipitation product is generated based on the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product, and the time-varying fusion weight.

[0012] In combination with the first aspect, optionally, a method for determining the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product includes:

[0013] Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ;

[0014] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3;

[0015] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the precipitation product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two precipitation products are set to 0;

[0016] If the observed precipitation value of a rain gauge is between any two values ​​of all precipitation products, the weights of the corresponding two precipitation products are calculated by the following formula, and the weights of the remaining precipitation products are 0:

[0017] ;

[0018] ;

[0019] in, is the observed precipitation value at the rain gauge station;

[0020] Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

[0021] In combination with the first aspect, optionally, the method for generating the time-varying fusion weight includes:

[0022] At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called;

[0023] Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

[0024] In combination with the first aspect, optionally, the time-varying fusion weight The calculation formula is:

[0025] .

[0026] In combination with the first aspect, optionally, the fused precipitation product is calculated using the following formula:

[0027]

[0028] Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

[0029] In a second aspect, the present invention provides a multi-source precipitation information time-varying fusion device considering error variation, comprising:

[0030] A precipitation product processing module is used to generate a first satellite inversion precipitation product, a second satellite inversion precipitation product and a reanalysis simulation precipitation product with the same temporal and spatial resolution;

[0031] A time-varying optimal weight generation module is used to determine the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product at each time point of each rainfall station with the goal of minimizing the error between the fused precipitation product and the rainfall station observation data;

[0032] A time-varying fusion weight generation module is used to interpolate the time-varying optimal weight using an inverse distance weighted method to generate a time-varying fusion weight, wherein the time-varying fusion weight has the same temporal and spatial resolution as the three precipitation products;

[0033] The fused precipitation product generation module is used to generate a fused precipitation product based on the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product, as well as the time-varying fusion weight.

[0034] In conjunction with the second aspect, optionally, the method for determining the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product includes:

[0035] Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ;

[0036] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3;

[0037] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the precipitation product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two precipitation products are set to 0;

[0038] If the observed precipitation value of a rain gauge is between any two values ​​of all precipitation products, the weights of the corresponding two precipitation products are calculated by the following formula, and the weights of the remaining precipitation products are 0:

[0039] ;

[0040] ;

[0041] in, is the observed precipitation value at the rain gauge station;

[0042] Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

[0043] In conjunction with the second aspect, optionally, the method for generating the time-varying fusion weight includes:

[0044] At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called;

[0045] Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

[0046] In combination with the second aspect, optionally, the time-varying fusion weight The calculation formula is:

[0047] ;

[0048] The fused precipitation product is calculated using the following formula:

[0049]

[0050] Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

[0051] In a third aspect, the present invention provides a multi-source precipitation information time-varying fusion system considering error variation, including a storage medium and a processor;

[0052] The storage medium is used to store instructions;

[0053] The processor is configured to operate according to the instructions to execute the method according to any one of the first aspects.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention makes full use of the advantages of satellite inversion, reanalysis simulation and ground rain gauge observation, overcomes the shortcomings of low performance of satellite inversion precipitation products and reanalysis simulation precipitation products and weak spatial representativeness of ground rain gauges, and obtains high-quality precipitation products through multi-source data time-varying fusion technology. The precipitation products provide more accurate and highly operational data support for water resources management and water disaster monitoring and early warning in data-free basins.

[0056] The present invention takes into account the daily temporal variability of precipitation and the time-varying characteristics of the fusion weights, thereby improving multiple performances of precipitation estimation, including correlation, systematic deviation, absolute difference and detection of rainfall, which is beneficial to water resources planning and management in data-free river basins and joint operation and scheduling of reservoir groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only 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 labor, among which:

[0058] Figure 1 A specific flow chart of the time-varying fusion method of multi-source precipitation information according to an embodiment of the present invention;

[0059] Figure 2 It is the feasible domain of weights w1 and w2 in the fusion method of the embodiment of the present invention and the possible optimal value of fusion precipitation. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0062] Example 1

[0063] The embodiment of the present invention provides a method for time-varying fusion of multi-source precipitation information considering error variation, comprising the following steps:

[0064] (1) Interpolate the original first satellite inverted precipitation product, the original second satellite inverted precipitation product and the original reanalysis simulated precipitation product to generate the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product with the same temporal and spatial resolution;

[0065] (2) At each time point of each rainfall station, the time-varying optimal weights of the first satellite-inverted precipitation product, the second satellite-inverted precipitation product, and the reanalysis simulated precipitation product are determined with the goal of minimizing the error between the fused precipitation product and the rainfall station observation data;

[0066] (3) using the inverse distance weighted method to interpolate the time-varying optimal weights to generate time-varying fusion weights, where the time-varying fusion weights have the same temporal and spatial resolution as the three precipitation products;

[0067] (4) generating a fused precipitation product based on the first satellite inverted precipitation product, the second satellite inverted precipitation product, the reanalysis simulated precipitation product, and the time-varying fusion weight.

[0068] In a specific implementation of the embodiment of the present invention, the method for determining the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product includes:

[0069] Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ;

[0070] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3;

[0071] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two products are set to 0;

[0072] If the observed precipitation value of a rain gauge is between any two values ​​of all product precipitation values, the weights of the corresponding two product precipitation values ​​are calculated by the following formula, and the weights of the remaining product precipitation values ​​are 0:

[0073] ;

[0074] ;

[0075] in, is the observed precipitation value at the rain gauge station;

[0076] Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

[0077] In a specific implementation of the embodiment of the present invention, the method for generating the second time-varying optimal weight matrix includes:

[0078] At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called;

[0079] Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

[0080] In a specific implementation of the embodiment of the present invention, the time-varying fusion weight The calculation formula is:

[0081] .

[0082] In a specific implementation of the embodiment of the present invention, the fused precipitation product is calculated using the following formula:

[0083]

[0084] Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

[0085] The time-varying fusion method of multi-source precipitation information considering error variation in an embodiment of the present invention is described in detail below in conjunction with a specific implementation manner.

[0086] refer to Figure 1 The present invention provides a time-varying fusion method for multi-source precipitation information considering error variation. The method first collects limited rain gauge observation data, satellite inversion precipitation products and reanalysis simulation precipitation products in data-deficient areas, and interpolates the three to have the same appropriate spatiotemporal resolution; secondly, at each time node of the rain gauge station, a linear combination method is applied to minimize the error between the fused precipitation value and the station observation data, and the weights of all time points are optimized to obtain the time-varying optimal weight; then the inverse distance weighted method is used to interpolate the time-varying optimal weight into a grid unit of an appropriate size, and the time-varying fusion weight of the grid is obtained through normalization processing; finally, at the same spatiotemporal resolution, the time-varying fusion weight is linearly weighted with the three precipitation products to generate a higher-precision fused precipitation product.

[0087] The following are examples and the attached Figure 1 , the technical solution of the present invention is further described in detail:

[0088] Step 1: Collect limited rainfall station observation data, satellite-derived precipitation products, and reanalysis-simulated precipitation products in data-deficient areas;

[0089] In this embodiment, two satellite precipitation products, GPM IMERG Final and PERSIANN-CDR, and ERA5 reanalysis precipitation products are used. Since there is a problem of mismatch between the temporal and spatial scales of these three precipitation products, bilinear interpolation is used to make them have the same appropriate temporal and spatial resolution to facilitate the following calculations. Among them, this embodiment uses a spatial resolution of 0.25° and a daily time scale.

[0090] Step 2: At each time point of the rain gauge station, the linear combination method is applied to minimize the error between the fused precipitation product and the rain gauge station observation data, and the weights of all time points are optimized to obtain the time-varying optimal weights. The specific steps are as follows:

[0091] Without any assumptions, three weights (w1, w2, w3) with values ​​ranging from 0 to 1 and summing to 1 are assigned to each rain gauge, and the precipitation values ​​of the three precipitation products (x1, x2, x3) are linearly fused into a unique fused precipitation value (m) at each moment:

[0092] (1)

[0093] The absolute error (d) between the fused precipitation value and the precipitation value at the rain gauge (r) is calculated using the function It means that the goal of this function is to minimize the absolute error value by optimizing the weights, so as to obtain the optimal fused precipitation value. There are usually three situations when multi-source precipitation products estimate reference precipitation data at each moment: all product precipitation values ​​underestimate the reference precipitation value, which means that x1, x2 and x3 are all less than r; all product precipitation values ​​overestimate the reference precipitation value, which means that any value of x1, x2 and x3 is greater than r; the three product precipitation values ​​underestimate and overestimate the reference precipitation value, which means that the r value is between the minimum and maximum values ​​of x1, x2 and x3.

[0094] In order to find the optimal weights in these three cases, it is necessary to determine the numerical range of the fused precipitation value. According to the constraint of the sum of weights, any weight can be expressed by the other two weights, such as , then equation (1) can be updated as:

[0095] (2)

[0096] Combined with the graphical method of linear programming, the feasible domain of weights w1 and w2 can be inferred, such as Figure 2 . From this figure, we can see that the feasible domain of these two weights is a triangular convex set, which means that the three vertices of the triangle are the optimal solutions of equation (2). Regardless of the direction and magnitude of the linear slope in equation (2), the possible optimal values ​​of fused precipitation are x1 (w1 is 1 and w2 is 0), x2 (w1 is 0 and w2 is 1), and x3 (w1 and w2 are both 0). In short, the range of fused precipitation values ​​is from the minimum to the maximum of x1, x2, and x3. Therefore, when all product precipitation values ​​are lower or higher than the reference precipitation value, the optimal weight of the product precipitation value closest to the reference precipitation value is 1, while the optimal weights of other product precipitation values ​​are 0.

[0097] For the reference precipitation value within the numerical range of the fused precipitation value, the absolute error value can reach the most ideal value, that is, the d value is 0. However, in this case, equation (2) has infinite solutions for the weights. Two product precipitation values ​​closer to the reference precipitation value can be used to calculate the fused precipitation value, but the reference precipitation value must be between these two product precipitation values ​​(such as x1 and x2). The weights of the remaining product precipitation values ​​(such as x3) are set to 0 to remove redundant precipitation information. Therefore, let the fused precipitation value be equal to the reference precipitation value, and the corresponding absolute error value is 0. Equation (2) can be simplified to:

[0098] (3)

[0099] Combined with the constraints of equation (3) , the expression of the optimal weight can be written as:

[0100] , (4)

[0101] In summary, the brief process of obtaining the optimal weight at each moment is as follows: if the precipitation values ​​of all original precipitation products are zero, their weights are all 1 / 3; if all product precipitation values ​​are lower or higher than the reference precipitation value, the weight of the product precipitation value closest to the reference precipitation value is 1, while the weights of other product precipitation values ​​are 0; if the reference precipitation value is between any two values ​​of all product precipitation values, the weights of the corresponding two product precipitation values ​​are calculated by equation (4), and the weights of the remaining product precipitation values ​​are 0. In this way, the time-varying optimal weight of the rainfall station can be obtained.

[0102] Step 3: Use the inverse distance weighted method to interpolate the time-varying optimal weights into grid cells of appropriate size (converting point data into grid data), and obtain the time-varying fusion weights of the grid through normalization.

[0103] At each moment, the Python tool in ArcGIS software is used to program batch processing of the time-varying optimal weights of the sites and interpolate them into grid weights of a specific grid cell size. In the batch processing process, the inverse distance weighted interpolation toolbox in ArcGIS software is called. After that, the grid weight is processed by normalization method to obtain the time-varying fusion weight of the grid. :

[0104] (5)

[0105] Step 4: At the same spatial and temporal resolution, the time-varying fusion weights are linearly weighted with the original precipitation product to generate a higher-precision fused precipitation product.

[0106] Based on three precipitation products The time-varying fusion weights of the grid are used to generate a fused precipitation product with temporal and spatial resolution. (0.25° / 1 day):

[0107] (6) .

[0108] Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

[0109] Example 2

[0110] Based on the same inventive concept as that of Example 1, an embodiment of the present invention provides a multi-source precipitation information time-varying fusion device considering error variation, comprising:

[0111] A precipitation product processing module is used to generate a first satellite inversion precipitation product, a second satellite inversion precipitation product and a reanalysis simulation precipitation product with the same temporal and spatial resolution;

[0112] A time-varying optimal weight generation module is used to determine the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product at each time point of each rainfall station with the goal of minimizing the error between the fused precipitation product and the rainfall station observation data;

[0113] A time-varying fusion weight generation module is used to interpolate the time-varying optimal weight using an inverse distance weighted method to generate a time-varying fusion weight, wherein the time-varying fusion weight has the same temporal and spatial resolution as the three precipitation products;

[0114] The fused precipitation product generation module is used to generate a fused precipitation product based on the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product, as well as the time-varying fusion weight.

[0115] In a specific implementation of the embodiment of the present invention, the method for determining the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product includes:

[0116] Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ;

[0117] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3;

[0118] If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the precipitation product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two precipitation products are set to 0;

[0119] If the observed precipitation value of a rain gauge is between any two values ​​of all precipitation products, the weights of the corresponding two precipitation products are calculated by the following formula, and the weights of the remaining precipitation products are 0:

[0120] ;

[0121] ;

[0122] in, is the observed precipitation value at the rain gauge station;

[0123] Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

[0124] In a specific implementation of the embodiment of the present invention, the method for generating the time-varying fusion weight includes:

[0125] At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called;

[0126] Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

[0127] In a specific implementation of the embodiment of the present invention, the time-varying fusion weight The calculation formula is:

[0128] ;

[0129] The fused precipitation product is calculated using the following formula:

[0130]

[0131] Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

[0132] Example 3

[0133] An embodiment of the present invention provides a multi-source precipitation information time-varying fusion system considering error variation, including a storage medium and a processor;

[0134] The storage medium is used to store instructions;

[0135] The processor is configured to operate according to the instructions to execute the method according to any one of the embodiments 1.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

[0141] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A time-varying fusion method for multi-source precipitation information considering error variation, characterized in that: include: Generate the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product with the same temporal and spatial resolution; At each time point of each rain gauge station, the time-varying optimal weights of the first satellite-inverted precipitation product, the second satellite-inverted precipitation product and the reanalysis simulated precipitation product are determined with the goal of minimizing the error between the fused precipitation product and the rain gauge station observation data; Using an inverse distance weighted method to interpolate the time-varying optimal weights to generate time-varying fusion weights, wherein the time-varying fusion weights have the same temporal and spatial resolution as the three precipitation products; A fused precipitation product is generated based on the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product, and the time-varying fusion weight.

2. According to claim 1, a method for time-varying fusion of multi-source precipitation information considering error variation is characterized by: The method for determining the time-varying optimal weights of the first satellite inversion precipitation product, the second satellite inversion precipitation product and the reanalysis simulation precipitation product includes: Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ; If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3; If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the precipitation product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two precipitation products are set to 0; If the observed precipitation value of a rain gauge is between any two values ​​of all precipitation products, the weights of the corresponding two precipitation products are calculated by the following formula, and the weights of the remaining precipitation products are 0: ； ； in, is the observed precipitation value at the rain gauge station; Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

3. The method for time-varying fusion of multi-source precipitation information considering error variation according to claim 1 is characterized by: The method for generating the time-varying fusion weights includes: At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called; Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

4. The method for time-varying fusion of multi-source precipitation information considering error variation according to claim 3 is characterized by: The time-varying fusion weight The calculation formula is: 。 5. The method for time-varying fusion of multi-source precipitation information considering error variation according to claim 4 is characterized by: The fused precipitation product is calculated using the following formula: ； Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

6. A multi-source precipitation information time-varying fusion device considering error variation, characterized in that: include: A precipitation product processing module is used to generate a first satellite inversion precipitation product, a second satellite inversion precipitation product and a reanalysis simulation precipitation product with the same temporal and spatial resolution; A time-varying optimal weight generation module is used to determine the time-varying optimal weights of the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product at each time point of each rainfall station with the goal of minimizing the error between the fused precipitation product and the rainfall station observation data; A time-varying fusion weight generation module is used to interpolate the time-varying optimal weight using an inverse distance weighted method to generate a time-varying fusion weight, wherein the time-varying fusion weight has the same temporal and spatial resolution as the three precipitation products; The fused precipitation product generation module is used to generate a fused precipitation product based on the first satellite inverted precipitation product, the second satellite inverted precipitation product and the reanalysis simulated precipitation product, as well as the time-varying fusion weight.

7. The device for time-varying fusion of multi-source precipitation information considering error variation according to claim 6 is characterized in that: The method for determining the time-varying optimal weights of the first satellite inversion precipitation product, the second satellite inversion precipitation product and the reanalysis simulation precipitation product includes: Assign three weights w1, w2, and w3 with a value range of 0 to 1 and a sum of 1 to each time point of each rainfall station. At each time point, the precipitation values ​​x1, x2, and x3 of the first satellite inverted precipitation product, the second satellite inverted precipitation product, and the reanalysis simulated precipitation product are linearly fused into a unique fused precipitation value m. The calculation formula of the fused precipitation value m is: ; If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product, and the reanalysis simulated precipitation product are all zero, then w1, w2, and w3 are all taken as 1 / 3; If the precipitation values ​​of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are all lower or higher than the observed precipitation value of the rain gauge station, the weight of the precipitation value of the precipitation product closest to the observed precipitation value of the rain gauge station is set to 1, and the weights of the precipitation values ​​of the other two precipitation products are set to 0; If the observed precipitation value of a rain gauge is between any two values ​​of all precipitation products, the weights of the corresponding two precipitation products are calculated by the following formula, and the weights of the remaining precipitation products are 0: ； ； in, is the observed precipitation value at the rain gauge station; Based on the three weights of each rain gauge at each time point, the time-varying optimal weights of the first satellite-retrieved precipitation product, the second satellite-retrieved precipitation product and the reanalysis simulated precipitation product are generated.

8. The device for time-varying fusion of multi-source precipitation information considering error variation according to claim 6 is characterized in that: The method for generating the time-varying fusion weights includes: At each time point, the Python tool in ArcGIS software is used to batch process the time-varying optimal weights of the three precipitation products at each rain gauge station and interpolate them into grid weights of a specific grid cell size. , , Indicates the index of weight; in batch processing, the inverse distance weighted interpolation toolbox in ArcGIS software is called; Using normalization to process mesh weights , get the time-varying fusion weight of the grid .

9. The device for time-varying fusion of multi-source precipitation information considering error variation according to claim 8, characterized in that: The time-varying fusion weight The calculation formula is: ； The fused precipitation product is calculated using the following formula: ； Among them, when hour, represents the first satellite inverted precipitation product; when hour, represents the second satellite inverted precipitation product, when hour, Represents the reanalysis simulated precipitation product.

10. A multi-source precipitation information time-varying fusion system considering error variation, characterized in that: including storage media and processors; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the method according to any one of claims 1-5.

Images