Extreme rainfall spatial scale identification method and system

Abstract

The invention provides an extreme rainfall spatial scale identification method and system, and the method carries out the following steps for each point in a to-be-identified region: determining extreme rainfall threshold values of a designated point and adjacent points of the designated point in the current round; performing dual polarization processing on the rainfall data of the specified point and the adjacent points based on the extreme rainfall threshold values of the specified point and the adjacent points; obtaining measurement windows in different search directions; dividing each measurement window into a plurality of polygons; fitting and solving according to the average semi-variation function value of the polygon in each search direction to obtain a corresponding space correlation length; and calculating to obtain a precipitation spatial scale range of the specified point according to the spatial correlation lengths in all the search directions. According to the method, the spatial scale range of extreme rainfall can be identified more comprehensively, more scientifically and more accurately.

Description

Technical Field

[0001] The present invention belongs to the field of hydrological and climate technology, and in particular relates to a method and system for identifying the spatial scale of extreme precipitation. Background Art

[0002] The spatial scale of extreme precipitation, as an important component of the spatial dimension of precipitation, is a key link in understanding the characteristics of extreme precipitation events and the complexity of the phenomenon. At present, there are two main methods for identifying the spatial scale (or spatial length) of extreme precipitation, both of which essentially utilize the relationship between the spatial correlation of variables and the decay of distance:

[0003] The method based on the exponential model relies on time dimension information (for example, the length of the time series of the variable), mainly considers the overall correlation of the time series of the variable between different locations, and the calculation is relatively simple. For example, Smith et al. studied the local spatial scale of tropical precipitation systems based on TRMM satellite rainfall rate data using the exponential model method. The analysis found that when the rainfall threshold is 0.1 mm / h, the average spatial scale over land is 90.5 km, and the ocean is 122.3 km. When the rainfall threshold is 1 mm / h, the value is slightly higher.

[0004] The method based on the semivariogram emphasizes the spatial distribution characteristics of the variable, uses the spatial variance of the variable to represent the degree of spatial autocorrelation of the variable, and is more flexible in its reliance on the time dimension information of the variable. For example, Touma et al. quantified the length scale of extreme daily precipitation in the United States from 1965 to 2014 based on the binarization distribution of precipitation records of the station network (with the 90% quantile extreme daily precipitation as the threshold), using the annular neighborhood smoothing method and the semivariogram index, indicating that its length scale varies in both regions and seasons. In the eastern half of the United States, its length scale can reach 400 km in winter, but its value is about half of the former in summer. Similarly, Tan et al. used the restricted neighborhood method and the semivariogram index to identify the spatial range of global precipitation extreme events from 1983 to 2018, pointing out that in non-monsoon areas, the spatial range of precipitation extreme events in autumn and winter is greater than that in spring and summer, and the spatial range of annual average precipitation extreme events exceeds 500 km, which is greater than that in monsoon areas. In addition, Du et al. quantified the spatial scale of Meiyu heavy precipitation events in the Yangtze River Basin from a geometric perspective using a rotating caliper algorithm. They conducted a comprehensive analysis of different types of Meiyu heavy precipitation events and found that they have regular spatial scales with an average length of about 1400km, an average width of about 500km, and an average range of about 40×104km2, indicating that Meiyu heavy precipitation events have a similar size to the sub-synoptic scale Meiyu front (1500-2000km).

[0005] Although the current research on the spatial scale of extreme precipitation has laid a methodological foundation, there are two main problems with the existing identification methods: 1) the threshold-based binarization scheme fails to scientifically characterize the intrinsic relationship between the distribution of extreme precipitation intensity and the spatial scale; 2) the use of circular or annular neighborhood window design does not take into account the spatial heterogeneity of extreme precipitation, which is the key to analyzing the spatial scale structure of extreme precipitation. These problems indicate that the current understanding of the spatial scale of extreme precipitation is still insufficient, and its quantitative characterization needs to be further explored. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the above-mentioned background technology and to provide a method and system for identifying the spatial scale of extreme precipitation, which takes into account objective characteristics such as extreme precipitation intensity field and anisotropy, can identify the spatial correlation lengths of extreme precipitation in multiple directions, and can identify the spatial scale range of extreme precipitation in a more comprehensive, scientific and accurate manner, thus solving the problems of the lack of objectivity and refinement in the existing methods in describing the spatial scale of extreme precipitation.

[0007] The technical solution adopted by the present invention is: a method for identifying the spatial scale of extreme precipitation, performing the following steps for each point in the area to be identified:

[0008] Determine the extreme precipitation thresholds for the specified point and its neighboring points in this round;

[0009] Based on the extreme precipitation threshold of the specified point and its neighboring points, the precipitation data of the specified point and its neighboring points are polarized: the precipitation data greater than or equal to the corresponding extreme precipitation threshold remains unchanged, and the precipitation data less than the extreme precipitation threshold is set to 0;

[0010] Starting from the specified point, according to the set search radius, distance step, bandwidth and angle tolerance, the measurement windows in different search directions are obtained; each measurement window is divided into several polygons;

[0011] Calculate the average semivariogram value of each polygon in each search direction at the moment when the precipitation data of all specified points in the specified time period is greater than the extreme precipitation threshold; fit and solve the corresponding spatial correlation length based on the average semivariogram value of the polygon in each search direction;

[0012] The spatial scale range of precipitation at a specified point is calculated based on the spatial correlation length in all search directions.

[0013] In the above technical solution, the process of determining the extreme precipitation threshold of any point in the current round of designated points and its neighboring points includes: sorting the historical precipitation data of the point from small to large to obtain a precipitation sequence; in the precipitation sequence, selecting the precipitation data at a set percentile position as the extreme precipitation threshold of the point.

[0014] In the above technical solution, the process of determining the extreme precipitation threshold of the current round of designated points and any of its neighboring points also includes: before generating the precipitation sequence, removing precipitation data with precipitation less than the set value.

[0015] In the above technical solution, the number of polygons into which each measurement window is divided is a multiple of the search radius and the distance step length.

[0016] In the above technical solution, in the process of generating the measurement window, on the basis of setting the search radius, 2 distance steps are dynamically increased successively to expand the search radius.

[0017] In the above technical solution, the following formula is used to calculate the average semivariogram value of the moment when the precipitation data of all designated points in the i-th polygon in the specified time period is greater than the extreme precipitation threshold in the n-th search direction:

[0018]

[0019] Among them, MR represents the point pair formed by the specified point M and its adjacent point R, and the precipitation at point R is P r,t ≠0; MO represents the point pair formed by the spatial point M and its adjacent point O, and the precipitation at point O is P O,t =0; and They represent the number of MR point pairs and the number of MO point pairs in the i-th polygon at time t in the n-th search direction, respectively; γ M-R is the semivariogram value of a set of MR point pairs, γ M-R =0.5(P M,t -P R,t ) 2 , the value range is [0,(P M,t -P r ) 2 ]; γ M-O is the semivariogram value of a set of MO point pairs, γ M-O =0.5(P M,t -P O,t ) 2 , the value range is [0,(P M,t ) 2 ]; where T represents the set of moments when the precipitation data at a specified point in a specified time period is greater than the extreme precipitation threshold; P M,t represents the precipitation at the specified point M at time t; P R,t represents the precipitation at time point R; P O,t represents the precipitation at time point O; P r is the extreme precipitation threshold.

[0020] In the above technical solution, the following formula is used to calculate the precipitation spatial scale range S of a specified point:

[0021]

[0022] Where, d represents the number of search directions, j represents the number of search directions, and n j Indicates the jth search direction; CL ηj Represents the spatial correlation length of the j-th search direction.

[0023] The present invention provides an extreme precipitation spatial scale identification system, which is used to implement the extreme precipitation spatial scale identification method described in the above technical solution.

[0024] The present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for identifying the spatial scale of extreme precipitation described in the above technical solution is implemented.

[0025] The present invention also provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for identifying the spatial scale of extreme precipitation as described in the above technical solution by executing the computer instructions.

[0026] The beneficial effects of the present invention are as follows: the present invention integrates the objective attributes of extreme precipitation intensity field and anisotropy into the extreme precipitation spatial scale identification method, covering the entire process from data preprocessing to spatial scale range calculation, forming a complete extreme precipitation spatial scale identification system. The present invention performs polarization processing on precipitation data for each point to ensure that the analysis is focused on extreme precipitation events and avoids interference with the results by low precipitation. The present invention can capture the directional differences in spatial correlation and reflect the spatial heterogeneity of precipitation events through measurement window division and multi-directional analysis. The analysis of the present invention can be targeted at a specified time period and combined with actual application needs (such as extreme weather process analysis) to improve the timeliness and application value of the results.

[0027] Furthermore, the present invention specifies a method for determining an extreme precipitation threshold value, by sorting historical precipitation data and selecting precipitation data at a set percentile position as the threshold value; using the percentile method to determine the extreme precipitation threshold value, which has a statistical basis and can adapt to precipitation characteristics in different regions and time periods; determining the threshold value for each point separately, which can reflect the differences in precipitation characteristics in local areas and improve the accuracy of the threshold value; ensuring that the selection of precipitation threshold values ​​is focused on extreme precipitation events, providing a high-quality data basis for subsequent analysis.

[0028] Furthermore, before generating a precipitation sequence, the present invention needs to remove precipitation data that is less than a set value; by removing low precipitation data, interference is effectively reduced and the influence of noise on the calculation of extreme precipitation thresholds is avoided; the quality of precipitation sequences is improved to make threshold calculations more representative; the amount of meaningless data processing is reduced and calculation efficiency is improved.

[0029] Furthermore, the present invention stipulates that the number of polygons into which the measurement window is divided is a multiple of the search radius and the distance step; the measurement window is divided into polygons to clarify the units of spatial analysis, which is convenient for subsequent semivariogram function calculations; division according to a fixed step size ensures the precision of spatial analysis and can more accurately reflect changes in spatial correlations of precipitation; the number of divisions is directly related to the search radius and step size, and the analysis range and accuracy can be flexibly controlled by adjusting these parameters.

[0030] Furthermore, the present invention stipulates that when generating a measurement window, on the basis of the set search radius, two distance steps can be dynamically increased successively to expand the search radius; the dynamic expansion of the search radius allows the user to adjust the measurement window range according to analysis requirements; increasing the search radius enables the analysis to capture spatial correlations at farther distances and adapt to precipitation process studies at different scales; dynamic expansion helps to verify the rationality of the initial search radius and ensure the stability and accuracy of the analysis results.

[0031] Furthermore, the present invention provides a formula for calculating the average semivariogram value, clarifies the calculation rules for different point pairs, and a method for using a set of moments in a specified time period; the formula combines the relationship between the target point and the neighboring points, and can quantify the strength of the spatial correlation of precipitation; calculates the average value based on the specified time period, and can capture the changing characteristics of the spatial correlation of precipitation in different periods; distinguishes point pairs whose precipitation exceeds the threshold and does not exceed the threshold, so that the spatial correlation analysis is more comprehensive and can reflect the contribution of different types of point pairs.

[0032] Furthermore, the present invention provides a formula for calculating the spatial scale range of precipitation at a target point, which is obtained by comprehensive calculation of multi-directional spatial correlation lengths; the comprehensive multi-directional spatial correlation lengths can accurately describe the spatial extension range of precipitation events; the formula takes into account the differences in multi-directional correlation lengths and can reflect the anisotropic characteristics of precipitation events in space; the provided spatial scale range results provide quantitative indicators for disaster prevention and mitigation, climate research and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0034] Figure 2 It is a schematic diagram of a measurement window of a specific embodiment. DETAILED DESCRIPTION

[0035] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments to facilitate a clear understanding of the present invention, but they do not constitute a limitation on the present invention.

[0036] Example 1

[0037] like Figure 1 As shown, the present invention provides a spatial scale identification method for extreme precipitation, and the following steps are performed for each point in the area to be identified:

[0038] Determine the extreme precipitation thresholds for the specified point and its neighboring points in this round;

[0039] Based on the extreme precipitation threshold of the specified point and its neighboring points, the precipitation data of the specified point and its neighboring points are polarized: the precipitation data greater than or equal to the corresponding extreme precipitation threshold remains unchanged, and the precipitation data less than the extreme precipitation threshold is set to 0;

[0040] Starting from the specified point, according to the set search radius, distance step, bandwidth and angle tolerance, the measurement windows in different search directions are obtained; each measurement window is divided into several polygons;

[0041] Calculate the average semivariogram value of each polygon in each search direction at the moment when the precipitation data of all specified points in the specified time period is greater than the extreme precipitation threshold; fit and solve the corresponding spatial correlation length based on the average semivariogram value of the polygon in each search direction;

[0042] The spatial scale range of precipitation at a specified point is calculated based on the spatial correlation length in all search directions.

[0043] The principle of the present invention is further explained below in conjunction with specific data.

[0044] The present invention provides a method for identifying the spatial scale of extreme precipitation, and the following steps are performed for each point in the identification area:

[0045] S1: Determine extreme precipitation thresholds based on the percentile method:

[0046] S1.1, for the daily precipitation data of the specified point M during the study period, remove the data with precipitation less than 0.1 mm to obtain the cleaned precipitation sequence P o :

[0047] P o = {P lon,lat,t |P lon,lat,t ≥0.1},t∈Y (1)

[0048] Where (lon, lat) is the spatial longitude and latitude of any point M; Y is the research period, t is a certain moment; P lon,lat,t is the precipitation at point M at time t.

[0049] S1.2, for the cleaned precipitation data P o Sort them in ascending order to get the precipitation sequence P s ;

[0050] S1.3, select an appropriate percentile perc as the standard for extreme precipitation threshold, and find the value of perc in the sorted precipitation data (P s ) and take the precipitation value at that location as the threshold value P r .

[0051] S1.4, according to steps S1.1-1.3, obtain the extreme precipitation thresholds of other points adjacent to point M.

[0052] Preferably, the following method can be selected to obtain the neighboring points of point M:

[0053] Based on geographic distance: Taking point M as the center, screening other points within the set search radius R is applicable to the search radius R limitation in the technical solution.

[0054] Based on regular grid structure: For regular grid data, neighboring points are directly obtained based on the grid index to efficiently process regular grid precipitation data.

[0055] Based on direction and distance: Filter neighboring points based on direction η and distance R. Support multi-directional analysis and measurement window generation.

[0056] Based on spatial query: Use GIS tools or spatial databases to directly obtain neighboring points through range queries, which is suitable for large-scale data management.

[0057] Based on precipitation correlation: According to the correlation or similarity of precipitation data, the neighboring points associated with the precipitation pattern of point M are screened. As a supplementary means, the targeting of neighboring point screening can be improved.

[0058] In this embodiment, the TRMM (TMPA) Precipitation L31 day 0.25degree x 0.25degreeV7 (https: / / disc.gsfc.nasa.gov / datasets / TRMM_3B42_Daily_7 / summary) precipitation product is taken as an example, the research period is from January 1, 2000 to December 30, 2019, the longitude and latitude of the selected point are 115.125°E and 30.125°N, and the percentile is selected as 95%. Then the extreme precipitation threshold Pr of the 95% quantile of this point is 51.46mm.

[0059] The data sets used in the actual calculation process (whether site observations or grid data) generally cover the entire area to be measured (or the world), and the neighboring points can be determined by scanning the window based on the spatial position relationship. The selection of neighboring points can reduce the amount of data to improve calculation efficiency, and can also guide the selection of the search radius.

[0060] Step S2, based on the extreme precipitation threshold of the designated point and its neighboring points, polarization processing is performed on the precipitation data of the designated point and its neighboring points to obtain the extreme precipitation spatial intensity field.

[0061] The extreme precipitation spatial intensity field S is calculated using the following formula: lon,lat,t :

[0062]

[0063] Among them, P r is the extreme precipitation threshold.

[0064] In the specific implementation process, for point M, the daily precipitation greater than or equal to 51.46 mm during the study period is kept unchanged, and the daily precipitation less than 51.46 mm is assigned a value of 0. Other adjacent points are also processed according to formula (2).

[0065] Step S3, constructing a spatial correlation length measurement window and polygon that take into account anisotropy.

[0066] The shape parameters of the spatial correlation length measurement window include: search radius R = 500 km, distance step h = 25 km, bandwidth bandw = 56.95 km and angle tolerance α = π / 8 (i.e. 22.5°). Taking the east direction as 0° and π / 4 (i.e. 45°) as the interval, the spatial correlation lengths of extreme precipitation in 8 different search directions are considered. The 8 search directions η are: 0, π / 4, π / 2, 3π / 4, π, 5π / 4, 6π / 4 and 7π / 4.

[0067] For a specified point M(lon,lat), in the search direction η, the coordinates of the feature points in the spatial correlation length measurement window are: (lon,lat),

[0068] (lon+1.4883cos(η-α),lat+1.4883sin(η-α)),

[0069] (lon+5.1565cos(η-0.1107),lat+5.1565sin(η-0.1107)),

[0070] (lon+5.1565cos(η+0.1107),lat+5.1565sin(η+0.1107)),

[0071] (lon+1.4883cos(η+α),lat+1.4883sin(η+α)),

[0072] (lon,lat).

[0073] As needed, the search radius can be expanded by dynamically increasing the step size by 2 steps on the basic search radius R to obtain the measurement window in each search direction. The core purpose of expanding the search radius is to adjust the range of the measurement window in order to capture spatial correlations at longer distances.

[0074] Based on the search radius R, starting from the specified point M (lon, lat), along each search direction η, according to the distance step h, each measurement window is divided into a certain number of polygons, that is, the number of polygons g = R / h = 500km / 25km = 20.

[0075] The vertex coordinates of each polygon can be deduced from the feature point coordinates, distance step, etc. of the corresponding measurement window.

[0076] In this embodiment, taking the search direction of π / 2 (i.e., 90°) as an example, the spatial relationship between the spatial correlation length measurement window and the extreme precipitation intensity field at point M on May 26, 2000 is visualized, as shown in FIG. Figure 2 shown.

[0077] Step S4, obtaining a time set when the direct fixed-point precipitation is greater than the extreme precipitation threshold.

[0078] The set of moments T when the precipitation at a specified point is greater than the extreme precipitation threshold is as follows:

[0079] T={t|P lon,lat,t ≥P r} (3)

[0080] During the specific implementation process, during the study period, a total of 125 days were identified in which the precipitation at point M was greater than the extreme precipitation threshold (51.46 mm).

[0081] Step S5, calculating the spatial correlation lengths of multiple search directions of any point based on the semivariogram and the exponential model.

[0082] The following formula is used to calculate the average semivariogram value of the moment when the precipitation data of all specified points in the ith polygon in the specified time period is greater than the extreme precipitation threshold:

[0083]

[0084] Among them, MR represents the point pair formed by the specified point M and its adjacent point R, and the precipitation at point R is Pr,t ≠0; MO represents the point pair formed by the spatial point M and its adjacent point O, and the precipitation at point O is P O,t =0; and They represent the number of MR point pairs and the number of MO point pairs in the i-th polygon at time t in the n-th search direction, respectively; γ M-R is the semivariogram value of a set of MR point pairs, γ M-R =0.5(P M,t -P R,t ) 2 , the value range is [0,(P M,t -P r ) 2 ]; γ M-O is the semivariogram value of a set of MO point pairs, γ M-O =0.5(P M,t -P O,t ) 2 , the value range is [0,(P M,t ) 2 ]; where T represents the set of moments when the precipitation data at a specified point in a specified time period is greater than the extreme precipitation threshold; P M,t represents the precipitation at the specified point M at time t; P R,t represents the precipitation at time point R; P O,t represents the precipitation at time point O; P r is the extreme precipitation threshold.

[0085] The parameter a0 of the corresponding exponential model is obtained by fitting the average semivariogram function value of several polygons in the nth search direction based on the following equation:

[0086]

[0087] Among them, C0, C, and a0 represent the parameters of the exponential model, h i Represents the distance from the geometric center of the i-th polygon to point M.

[0088] The spatial correlation length CL along the search direction η is calculated using the following formula: η :

[0089] CL η =3a0 (6)

[0090] Where CLη represents the spatial correlation length of extreme precipitation at point M along the search direction η.

[0091] In the present embodiment, taking the search direction of π / 2 (i.e. 90°) as an example, the number of MR (MO) point pairs gradually decreases (increases) and then tends to be stable as the distance to point M increases. Using the exponential function to solve, we know that a0 = 102.48 km, so the spatial correlation length of the extreme precipitation at point M in the search direction of π / 2 (i.e. 90°) is CL π / 2 =3a0=307.43km. Calculating in sequence, the other search directions (0, π / 4, 3π / 4, π, 5π / 4, 6π / 4 and 7π / 4) are 269.54km, 469.69km, 352.15km, 328.89km, 273.35km, 264.66km and 339.98km respectively.

[0092] Step S6, identifying the spatial scale range of extreme precipitation at a specified point based on the multi-directional spatial correlation length.

[0093] The following formula is used to calculate the spatial scale range S of extreme precipitation:

[0094]

[0095] Wherein, d represents the number of search directions (d=1, 2, 3, ..., 8), and they are sorted in a counterclockwise direction with the east direction being 0°.

[0096] In this specific example, it can be concluded that the spatial scale of extreme precipitation at point M is 29.66×104km 2 .

[0097] Example 2

[0098] The present invention provides an extreme precipitation spatial scale identification system, which is used to implement the extreme precipitation spatial scale identification method described in the above technical solution, including: an extreme precipitation threshold determination module, a polarization processing module, a polygon acquisition module, a spatial correlation length calculation module, and a precipitation spatial scale range calculation module;

[0099] The extreme precipitation threshold determination module is used to determine the extreme precipitation threshold of the designated point and its neighboring points in this round;

[0100] The polarization processing module is used to perform polarization processing on the precipitation data of the designated point and its neighboring points based on the extreme precipitation threshold of the designated point and its neighboring points: the precipitation data greater than or equal to the corresponding extreme precipitation threshold remains unchanged, and the precipitation data less than the extreme precipitation threshold is set to 0;

[0101] The polygon acquisition module is used to obtain the measurement windows in different search directions starting from the specified point according to the set search radius, distance step, bandwidth and angle tolerance; each measurement window is divided into several polygons;

[0102] The spatial correlation length calculation module is used to calculate the average semivariogram value of each polygon in each search direction at the moment when the precipitation data of all designated points in the designated time period is greater than the extreme precipitation threshold; the corresponding spatial correlation length is obtained by fitting and solving the average semivariogram value of the polygon in each search direction;

[0103] The precipitation spatial scale range calculation module is used to calculate the precipitation spatial scale range of a specified point based on the spatial correlation length in all search directions.

[0104] Based on the first law of geography, the present invention fully considers the objectivity of precipitation intensity field and anisotropy while emphasizing the spatial distribution characteristics of precipitation, and uses the percentile threshold method to highlight the spatial type of extreme precipitation intensity field. Combined with the semivariogram function, the nonlinear relationship of spatial correlation of extreme precipitation intensity in different directions with distance attenuation is clarified, so that the spatial scale range of extreme precipitation can be identified more objectively and scientifically. It can provide technical support for the impact or risk assessment of extreme precipitation disasters, and can also provide theoretical method support for revealing the complex spatial characteristics of extreme precipitation. At the same time, it can provide a reference for the spatial scale identification of other hydrological and meteorological variables, and has good promotion significance.

[0105] Example 3

[0106] The present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for identifying the spatial scale of extreme precipitation described in the above technical solution is implemented.

[0107] Example 4

[0108] The present invention also provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for identifying the spatial scale of extreme precipitation as described in the above technical solution by executing the computer instructions.

[0109] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take 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 code.

[0110] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 produce 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.

[0111] 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.

[0112] 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.

[0113] 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 modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

[0114] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

Claims

1. A method for identifying the spatial scale of extreme precipitation, characterized by: Perform the following steps for each point in the area to be identified: Determine the extreme precipitation thresholds for the specified point and its neighboring points in this round; Based on the extreme precipitation threshold of the specified point and its neighboring points, the precipitation data of the specified point and its neighboring points are polarized: the precipitation data greater than or equal to the corresponding extreme precipitation threshold remains unchanged, and the precipitation data less than the extreme precipitation threshold is set to 0; Starting from the specified point, according to the set search radius, distance step, bandwidth and angle tolerance, the measurement windows in different search directions are obtained; each measurement window is divided into several polygons; Calculate the average semivariogram value of each polygon in each search direction at the moment when the precipitation data of all specified points in the specified time period is greater than the extreme precipitation threshold; The corresponding spatial correlation length is obtained by fitting the average semivariogram value of the polygon in each search direction; The spatial scale range of precipitation at a specified point is calculated based on the spatial correlation length in all search directions.

2. A method according to claim 1, characterized in that: The process of determining the extreme precipitation threshold of any point in the current round of designated points and its neighboring points includes: sorting the historical precipitation data of the point from small to large to obtain a precipitation sequence; in the precipitation sequence, selecting the precipitation data at the set percentile position as the extreme precipitation threshold of the point.

3. A method according to claim 2, characterized in that: The process of determining the extreme precipitation threshold of any point in the current round and its neighboring points also includes: before generating the precipitation sequence, removing precipitation data with precipitation less than a set value.

4. A method according to claim 1, characterized in that: The number of polygons into which each measurement window is divided is a multiple of the search radius and the distance step size.

5. A method according to claim 4, characterized in that: In the process of generating the measurement window, based on the set search radius, 2 distance steps are dynamically increased successively to expand the search radius.

6. A method according to claim 1, characterized in that: The following formula is used to calculate the average semivariogram value γ of the moment when the precipitation data of all specified points in the ith polygon in the specified time period is greater than the extreme precipitation threshold in the ηth search direction: i,η : Among them, MR represents the point pair formed by the specified point M and its adjacent point R, and the precipitation at point R is P r,t ≠0; MO represents the point pair formed by the spatial point M and its adjacent point O, and the precipitation at point O is P O,t =0; and They represent the number of MR point pairs and the number of MO point pairs in the i-th polygon at time t in the n-th search direction, respectively; γ M-R is the semivariogram value of a set of MR point pairs, γ M-R =0.5(P M,t -P R,t ) 2 , the value range is [0,(P M,t -P r ) 2 ]; γ M-O is the semivariogram value of a set of MO point pairs, γ M-O =0.5(P M,t -P O,t ) 2 , the value range is [0,(P M,t ) 2 ]; where T represents the set of moments when the precipitation data at a specified point in a specified time period is greater than the extreme precipitation threshold; P M,t represents the precipitation at the specified point M at time t; P R,t represents the precipitation at time point R; P O,t represents the precipitation at time point O; P r is the extreme precipitation threshold.

7. A method according to claim 1, characterized in that: The precipitation spatial scale range S at a specified point is calculated using the following formula: Where, d represents the number of search directions, j represents the number of search directions, and n j Indicates the jth search direction; CL ηj Represents the spatial correlation length of the j-th search direction.

8. An extreme precipitation spatial scale identification system, characterized in that: The system is used to implement the extreme precipitation spatial scale identification method described in any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the extreme precipitation spatial scale identification method described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the extreme precipitation spatial scale identification method according to any one of claims 1 to 7 by executing the computer instructions.

Images