Precipitation grading method for obtaining any time scale

Abstract

According to the precipitation grading method for obtaining any time scale, only 12-hour and 24-hour grading standards are given by the precipitation grading national standard (GB / T 28592-2012), and precipitation grading standards of other time scales are not given. However, many industries (such as traffic, hydrological monitoring and the like) possibly have requirements for precipitation division standards of other time scales (such as 10 minutes, 1 hour and the like), and in order to meet the requirements, an equal ranking rate method is provided, and the method can obtain precipitation division standards of any time scale not larger than the time resolution of precipitation data. Taking ERA5 hourly rainfall data and time scale 24-hour rainfall division standards (GB / T28592-2012) of Nanjing region (111 DEG E, 32 DEG N) as examples, the rainfall division standards of time scales of 1 hour, 3 hours, 6 hours and 12 hours are obtained by using the method, and the rainfall division standards of time scales of 12 hours obtained by using the method are close to the national standard. Therefore, if minute-level automatic station rainfall data is taken as a research object, the method can obtain a rainfall division standard of any time scale.

Description

Technical Field

[0001] This invention relates to the requirements of different business operations for the classification of precipitation levels at different time scales. Specifically, it proposes a method for obtaining the classification standard of precipitation levels at any time scale, and applies it to the 1-hour, 3-hour, 6-hour, and 12-hour precipitation level classification standards in Nanjing area (119°E, 32°N) as an example analysis. Background Technology

[0002] Currently, the national standard for precipitation classification (GB / T 28592-2012) applies to all regions of China, but it only provides classification standards for 12-hour and 24-hour periods, without providing standards for other time scales. However, many industries (such as transportation and hydrological monitoring) may have a need for precipitation classification standards at other time scales (such as 10-minute or 1-hour periods). To address this issue, this invention proposes a method for obtaining precipitation classification standards at any time scale, no greater than the temporal resolution of precipitation data. If minute-level automatic weather station precipitation historical data is used as the research object, precipitation classification standards at any time scale can be obtained, thereby meeting the needs of different industries for precipitation classification standards at different time scales. Summary of the Invention

[0003] The main contents of this invention include the following aspects:

[0004] a) A method for obtaining precipitation classification standards at any time scale is proposed, namely the "equi-ranking rate method", which aims to meet the needs of different industries for precipitation classification standards at different time scales.

[0005] b) In the equal ranking rate method, for the same region, it is assumed that the probability of precipitation level occurring at different time scales is the same. That is, it is believed that the probability of precipitation levels such as "light rain", "moderate rain" and "heavy rain" occurring in the same region is the same at different time scales such as the 24-hour time scale, the 1-hour time scale, and the 10-minute time scale.

[0006] c) According to the national standard (GB / T 28592-2012) for classifying 12-hour or 24-hour precipitation, by arranging the 12-hour or 24-hour precipitation from smallest to largest or from largest to smallest, the ranking rate of different precipitation levels can be obtained (i.e.: ).

[0007] d) Obtain precipitation data for the required time scale, arrange them from smallest to largest or from largest to smallest, and use the total sample length to multiply by the ranking rate of different precipitation levels based on 12 hours or 24 hours to obtain the position of different precipitation levels at the required time scale. The precipitation corresponding to this position is the upper limit of the classification standard for different precipitation levels at the required time scale. Attached Figure Description

[0008] Figure 1 This study investigates the variation characteristics of 24-hour precipitation over a timescale in the Nanjing area (119°E, 32°N). Hourly precipitation and snowfall data from the past 20 years (2005-2024) are used from the ERA5 dataset (175,320 samples). The dataset is named "ERA5 hourly data on single levels from 1940 to present" and can be found at: https: / / cds.climate.copernicus.eu / datasets / reanalysis-era5-single-levels?tab=download. The variable name for precipitation is "Total precipitation" (unit: m), and the variable name for snowfall is "Snowfall" (unit: mof water equivalent). Both precipitation and snowfall data have a horizontal resolution of 0.25° × 0.25° and a temporal resolution of 1 hour (unit: m), which are converted to mm before use.

[0009] Using hourly precipitation data from the past 20 years (2005-2024), 24-hour precipitation data can be obtained. After removing snowfall and no-precipitation data, the total number of 24-hour precipitation samples is 4305. These samples are sorted by precipitation amount from smallest to largest as follows: Figure 1 As shown, the rainfall value ranges from 4.77 × 10⁻⁶. -4 The rainfall was approximately 166.57 mm, with an average of 5.45 mm. According to the national standard (GB / T 28592-2012), the upper limits for 24-hour light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain are 9.9 mm, 24.9 mm, 49.9 mm, 99.9 mm, and 249.9 mm, respectively. This yields corresponding ranking rates of 83.51% for light rain, 94.70% for moderate rain, 98.70% for heavy rain, 99.95% for torrential rain, and 100.00% for extremely heavy rain.

[0010] Figure 2This describes the increasing variation of rainfall over 1 hour, 3 hours, 6 hours, and 12 hours in the Nanjing area (119°E, 32°N). Using ERA5 hourly precipitation and snowfall data from the past 20 years (2005-2024), precipitation data at any integer multiple of 1 hour can be obtained. For example, after removing snowfall and no-precipitation data from 1-hour, 3-hour, 6-hour, and 12-hour precipitation data, the total number of rainfall samples for 1-hour, 3-hour, 6-hour, and 12-hour timescales are 49964, 20041, 11745, and 7204, respectively. The variation graph after ranking the rainfall from smallest to largest is shown below. Figure 2 As shown, the rainfall values ​​ranged from 2.38 × 10⁻⁶. -4 ~18.94mm, 2.38×10 -4 ~49.35mm, 2.38×10 -4 ~80.05mm, 4.77×10 -4 The diameters range from 150.38 mm to 150.38 mm, with average values ​​of 0.47 mm, 1.17 mm, 2.00 mm, and 3.25 mm, respectively.

[0011] Figure 3 These are the upper limits of the standards for classifying different rainfall levels at time scales of 1 hour, 3 hours, 6 hours, and 12 hours in the Nanjing area (119°E, 32°N). Using the equal ranking rate method, the ranking rates for light rain, moderate rain, heavy rain, rainstorm, and torrential rain at time scales of 1 hour, 3 hours, 6 hours, and 12 hours are 83.51%, 94.70%, 98.70%, 99.95%, and 100.00%, respectively. Based on this ranking rate and... Figure 2 Ranking the rainfall values ​​yields the upper limit values ​​for classifying light rain, moderate rain, heavy rain, rainstorm, and torrential rain at time scales of 1 hour, 3 hours, 6 hours, and 12 hours. Figure 3 ),exist Figure 3 To maintain consistency with the numerical precision of the national standard (GB / T 28592-2012), one decimal place is retained here. Figure 3 (a) shows that the upper limits of the standards for classifying light rain, moderate rain, heavy rain, rainstorm, and torrential rain at a time scale of 1 hour are 0.8 mm, 2.2 mm, 4.5 mm, 13.3 mm, and 18.9 mm, respectively. That is, the corresponding standards for light rain, moderate rain, heavy rain, rainstorm, and torrential rain are "light rain ≤ 0.8 mm", "0.8 < moderate rain ≤ 2.2 mm", "2.2 < heavy rain ≤ 4.5 mm", "4.5 < rainstorm ≤ 13.3 mm", "13.3 < torrential rain ≤ 18.9 mm", and "18.9 < extremely heavy rain". Figure 3(b) It can be seen that the upper limits of the standards for classifying light rain, moderate rain, heavy rain, rainstorm, and torrential rain at a time scale of 3 hours are 2.0 mm, 5.5 mm, 11.3 mm, 33.1 mm, and 49.3 mm, respectively. That is, the corresponding standards for light rain, moderate rain, heavy rain, rainstorm, and torrential rain are "light rain ≤ 2.0 mm", "2.0 < moderate rain ≤ 5.5 mm", "5.5 < heavy rain ≤ 11.3 mm", "11.3 < rainstorm ≤ 33.1 mm", "33.1 < torrential rain ≤ 49.3 mm", and "49.3 < extremely heavy rain". From... Figure 3 (c) shows that the upper limits of the standards for classifying light rain, moderate rain, heavy rain, rainstorm, and torrential rain on a 6-hour timescale are 3.5 mm, 9.4 mm, 19.3 mm, 61.1 mm, and 80.0 mm, respectively. That is, the corresponding standards for light rain, moderate rain, heavy rain, rainstorm, and torrential rain are "light rain ≤ 3.5 mm", "3.5 < moderate rain ≤ 9.4 mm", "9.4 < heavy rain ≤ 19.3 mm", "19.3 < rainstorm ≤ 61.1 mm", "61.1 < torrential rain ≤ 80.0 mm", and "80.0 < extremely heavy rain". Figure 3 (d) It can be seen that the upper limits of the standards for classifying light rain, moderate rain, heavy rain, rainstorm, and torrential rain at a time scale of 12 hours are 5.8 mm, 15.3 mm, 31.2 mm, 81.3 mm, and 150.4 mm, respectively. That is, the corresponding standards for light rain, moderate rain, heavy rain, rainstorm, and torrential rain are "light rain ≤ 5.8 mm", "5.8 < moderate rain ≤ 15.3 mm", "15.3 < heavy rain ≤ 31.2 mm", "31.2 < rainstorm ≤ 81.3 mm", "81.3 < torrential rain ≤ 150.4 mm", and "150.4 < extremely heavy rain".

[0012] Table 1 shows the rainfall levels in the Nanjing area (119°E, 32°N) at 1-hour, 3-hour, 6-hour, and 12-hour time scales (unit: mm).

[0013] Detailed Implementation

[0014] For historical precipitation data of a certain region at a time scale T (unit: hour), the total number of samples is N (N≥24), where the value 12 is a multiple of the time scale T (the multiple is an integer greater than 1).

[0015] Suppose we want to obtain the classification standards for different precipitation levels on a time scale of M (unit: hour), where M is an integer multiple of T. That is, it satisfies the following relationship:

[0016] M = i × T (1)

[0017] In the above formula, i is a positive integer not less than 1.

[0018] Using historical precipitation data at time scale T, historical precipitation data at the required time scale M are obtained, with a total sample size of [missing information]. And round down to the nearest integer, denoted as Based on known records of manually observed weather phenomena, the number of samples with no precipitation or sleet was removed, leaving only the total number of samples with precipitation. All are the total number of samples of snowfall.

[0019] Will Sort the precipitation data from smallest to largest (or from largest to smallest) to obtain a new timescale M-hour precipitation array.

[0020] 1. Classification criteria based on different precipitation amounts over a 24-hour time scale

[0021] Using historical precipitation data at a time scale of T (unit: hour), historical precipitation data at a time scale of 24 hours can be obtained, along with its sample size. And round down to the nearest integer, denoted as Based on known records of manually observed weather phenomena, the number of samples with no precipitation or sleet was removed, leaving only the total number of samples with precipitation. All are the total number of samples of snowfall.

[0022] Will Sort the precipitation data from smallest to largest (or from largest to smallest) to obtain a new 24-hour precipitation array for a new time scale.

[0023] 1.1 Classification Standard of 24-hour Rainfall Based on National Standard (GB / T 28592-2012)

[0024] Table 1. Classification Standards for 24-Hour Rainfall in National Standard (GB / T 28592-2012)

[0025] standard Light rain Moderate rain heavy rain rainstorm Heavy rain Extreme rainstorm Rainfall / mm 0.1～9.9 10.0～24.9 25.0～49.9 50.0～99.9 100.0～249.9 ≥250.0

[0026] Based on Table 1 and the 24-hour rainfall data array arranged from smallest to largest, the positions corresponding to the values ​​closest to 9.9mm, 24.9mm, 49.9mm, 99.9mm, and 249.9mm were identified and denoted as follows: Get Xiaoyu's ranking rate moderate rain ranking Heavy rain ranking rate Rainstorm Ranking Heavy rain ranking The calculation expression is as follows:

[0027]

[0028]

[0029] According to the ranking rate method, using the national standard (GB / T 28592-2012) for 24-hour precipitation classification, the ranking rates of different levels of precipitation are independent of the time scale. That is, the ranking rate values ​​for light rain, moderate rain, heavy rain, rainstorm, and torrential rain at any time scale are all [values ​​to be filled in].

[0030] For an array of hourly rainfall data sorted from smallest to largest over a timescale of M, the number of samples is... According to the equal ranking rate method, the ranking rate values ​​corresponding to light rain, moderate rain, heavy rain, rainstorm, and torrential rain at a time scale of M hours are respectively... Using equations (2)-(6) above, we can find the location of light rain in the precipitation array with a time scale of M hours. Moderate rain corresponding location Location of heavy rain Location of heavy rain Location of heavy rain The calculation expression is as follows:

[0031]

[0032] Due to the results obtained from equations (7)-(11) The values ​​may have decimal parts, but their positions in the array should be integers. Therefore, a rounding method is used to obtain their corresponding integer positions. Specifically, the positions of light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain in the time-scale M-hour rainfall array, sorted from smallest to largest, are as follows: Knowing the positions of light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain in the array, we can directly obtain the corresponding values ​​at those positions, denoted as follows: This allows us to derive the criteria for classifying light rain on a time scale M. Standards for classifying moderate rain Heavy Rain Classification Standards Rainstorm Classification Standards Criteria for classifying heavy rainstorms Special Rainstorm Classification Standards They are in the following forms respectively:

[0033]

[0034] In the above formulas (12)-(17) The standard values ​​for classifying light rain, moderate rain, heavy rain, rainstorm, torrential rain, and extremely heavy rain at the required time scale M are given.

[0035] 1.2 24-hour snowfall classification standard based on national standard (GB / T 28592-2012)

[0036] Table 2. Standards for classifying 24-hour snowfall according to national standard (GB / T 28592-2012)

[0037] standard Light Snow Moderate snow heavy snow Blizzard Blizzard Massive blizzard Snowfall / mm 0.1～2.4 2.5～4.9 5.0～9.9 10.0～19.9 20.0～29.9 ≥30.0

[0038] Based on Table 2 and the 24-hour snowfall data arranged in ascending order, the positions corresponding to the values ​​closest to 2.4mm, 4.9mm, 9.9mm, 19.9mm, and 29.9mm were identified and denoted as follows: Get Xiaoxue's ranking rate Medium snow ranking rate Ranking rate during heavy snow Blizzard ranking rate Blizzard ranking The calculation expression is as follows:

[0039]

[0040] According to the equal ranking rate method, using the national standard (GB / T 28592-2012) for classifying 24-hour snowfall, we can obtain...

[0041] The ranking rates for different levels of precipitation are independent of the time scale; that is, the ranking rate values ​​for light snow, moderate snow, heavy snow, blizzard, and severe blizzard at any time scale are all respectively...

[0042] For an array of snowfall and precipitation data with a time scale of M hours, sorted from smallest to largest, the number of samples is: According to the equal ranking rate method, the ranking rate values ​​for light snow, moderate snow, heavy snow, blizzard, and severe blizzard at a time scale of M hours are respectively... Using equations (18)-(22) above, we can find the position of the light snow in the snowfall and precipitation array with a time scale of M hours. The corresponding location of moderate snow Location corresponding to heavy snow Blizzard's corresponding location Location of the blizzard The calculation expression is as follows:

[0043]

[0044] Due to the results obtained from equations (23)-(27) The values ​​may have decimal parts, but their positions in the array should be integers. Therefore, a rounding method is used to obtain their corresponding integer positions. Specifically, the positions of light snow, moderate snow, heavy snow, blizzard, and severe blizzard in the snowfall array sorted from smallest to largest over a time scale of M hours are as follows: Knowing the positions of Light Snow, Moderate Snow, Heavy Snow, Blizzard, and Major Blizzard in the array, we can directly obtain the values ​​corresponding to those positions, denoted as follows: From this, we can obtain the criteria for dividing Minor Snow on the time scale M. Standards for classifying moderate snow Heavy Snow Classification Standards Blizzard Classification Standards Blizzard Classification Standards Criteria for classifying extreme blizzards They are in the following forms respectively:

[0045]

[0046]

[0047] In the above formulas (28)-(33) The standard values ​​for classifying light snow, moderate snow, heavy snow, blizzard, severe blizzard, and extremely severe blizzard for the required time scale M.

[0048] 2. Classification criteria based on different precipitation amounts over a 12-hour time scale

[0049] Using historical precipitation data at a time scale of T (unit: hour), historical precipitation data at a time scale of 12 hours can be obtained, along with its sample size. Round down to the nearest integer, and the total number of samples is denoted as . Based on known records of manually observed weather phenomena, the number of samples with no precipitation or sleet was removed, leaving only the total number of samples with precipitation. All are the total number of samples of snowfall.

[0050] Will Sort the precipitation data from smallest to largest (or from largest to smallest) to obtain a new 12-hour precipitation array.

[0051] 2.1 Classification Standard of 12-hour Rainfall Based on National Standard (GB / T 28592-2012)

[0052] Table 3. Standards for classifying 12-hour rainfall according to national standard (GB / T 28592-2012)

[0053] standard Light rain Moderate rain heavy rain rainstorm Heavy rain Extreme rainstorm Rainfall / mm 0.1～4.9 5.0～14.9 15.0～29.9 30.0～69.9 70.0～139.9 ≥140.0

[0054] Based on Table 3 and the 12-hour rainfall data arranged from smallest to largest, the positions corresponding to the values ​​closest to 4.9mm, 14.9mm, 29.9mm, 69.9mm, and 139.9mm were identified and denoted as follows: Get Xiaoyu's ranking rate moderate rain ranking Heavy rain ranking rate Rainstorm Ranking Heavy rain ranking The calculation expression is as follows:

[0055]

[0056]

[0057] According to the ranking rate method, using the national standard (GB / T 28592-2012) for classifying 12-hour rainfall, the ranking rates of different levels of rainfall are independent of the time scale. That is, the ranking rate values ​​for light rain, moderate rain, heavy rain, rainstorm, and torrential rain at any time scale are all [values ​​to be filled in].

[0058] For an array of hourly rainfall data sorted from smallest to largest over a timescale of M, the number of samples is... According to the equal ranking rate method, the ranking rate values ​​corresponding to light rain, moderate rain, heavy rain, rainstorm, and torrential rain at a time scale of M hours are respectively... Using equations (34)-(38) above, we can find the location of light rain in the rainfall array with a time scale of M hours. Moderate rain corresponding location Location of heavy rain Location of heavy rain Location of heavy rain The calculation expression is as follows:

[0059]

[0060] Due to the results obtained from equations (39)-(43) The values ​​may have decimal parts, but their positions in the array should be integers. Therefore, a rounding method is used to obtain their corresponding integer positions. Specifically, the positions of light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain in the time-scale M-hour rainfall array, sorted from smallest to largest, are as follows: Knowing the positions of light rain, moderate rain, heavy rain, torrential rain, and extremely heavy rain in the array, we can directly obtain the corresponding values ​​at those positions, denoted as follows: This allows us to derive the criteria for classifying light rain on a time scale M. Standards for classifying moderate rain Heavy Rain Classification Standards Rainstorm Classification Standards Criteria for classifying heavy rainstorms Special Rainstorm Classification Standards They are in the following forms respectively:

[0061]

[0062] In the above equations (44)-(49) The standard values ​​for classifying light rain, moderate rain, heavy rain, rainstorm, torrential rain, and extremely heavy rain at the required time scale M are given.

[0063] 2.2 12-hour snowfall classification standard based on national standard (GB / T 28592-2012)

[0064] Table 4. Standards for classifying 12-hour snowfall according to national standard (GB / T 28592-2012)

[0065] standard Light Snow Moderate snow heavy snow Blizzard Blizzard Massive blizzard Snowfall / mm 0.1～0.9 1.0～2.9 3.0～5.9 6.0～9.9 10.0～14.9 ≥15.0

[0066] Based on Table 4 and the 12-hour snowfall data arranged from smallest to largest, the positions corresponding to the values ​​closest to 0.9mm, 2.9mm, 5.9mm, 9.9mm, and 14.9mm were identified and denoted as follows: Get Xiaoxue's ranking rate Medium snow ranking rate Ranking rate during heavy snow Blizzard ranking rate Blizzard ranking The calculation expression is as follows:

[0067]

[0068]

[0069] According to the ranking rate method, using the national standard (GB / T 28592-2012) for classifying 12-hour snowfall and precipitation, the ranking rates of different levels of precipitation are independent of the time scale. That is, the ranking rate values ​​for light snow, moderate snow, heavy snow, blizzard, and severe blizzard at any time scale are all respectively...

[0070] For an array of snowfall amounts in M ​​hours sorted from smallest to largest, the number of samples is... According to the equal ranking rate method, the ranking rate values ​​for light snow, moderate snow, heavy snow, blizzard, and severe blizzard at a time scale of M hours are respectively... Using equations (50)-(54) above, we can find the position of light snow in the snowfall data array with a time scale of M hours. The corresponding location of moderate snow Location corresponding to heavy snow Blizzard's corresponding location Location of the blizzard The calculation expression is as follows:

[0071]

[0072] Because of equations (55)-(59) The values ​​may have decimal parts, but their positions in the array should be integers. Therefore, a rounding method is used to obtain their corresponding integer positions. Specifically, the positions of light snow, moderate snow, heavy snow, blizzard, and severe blizzard in the snowfall array sorted from smallest to largest over a time scale of M hours are as follows: Knowing the positions of Light Snow, Moderate Snow, Heavy Snow, Blizzard, and Major Blizzard in the array, we can directly obtain the values ​​corresponding to those positions, denoted as follows: From this, we can obtain the criteria for dividing Minor Snow on the time scale M. Standards for classifying moderate snow Heavy Snow Classification Standards Blizzard Classification Standards Blizzard Classification Standards Criteria for classifying extreme blizzards They are in the following forms respectively:

[0073]

[0074] In the above formulas (60)-(65) The standard values ​​for classifying light snow, moderate snow, heavy snow, blizzard, severe blizzard, and extremely severe blizzard for the required time scale M.

Claims

1. A method for obtaining precipitation amount classification at any time scale, namely "equal-rank-rate" method, characterized in that, For any region, the ranking rates of different precipitation classification standards (such as "light rain", "moderate rain", "heavy rain", etc.) are obtained according to the 12-hour or 24-hour precipitation classification standard of the national standard (GB / T 28592-2012), and it is assumed that the ranking rates of different precipitation classification standards of different time scales (such as 10 minutes, 1 hour, 3 hours, etc.) in the same region are equal to the ranking rates obtained by the 12-hour or 24-hour precipitation classification standard, so as to obtain the precipitation classification standards of different time scales by using the ranking rate method.

2. The equal ranking rate method according to claim 1, characterized in that the ranking rates of different precipitation classification standards are obtained according to the 12-hour precipitation classification standard of the national standard (GB / T 28592-2012), so as to obtain the precipitation classification standards of different time scales.

3. The equal ranking rate method according to claim 1, characterized in that the ranking rates of different precipitation classification standards are obtained according to the 24-hour precipitation classification standard of the national standard (GB / T 28592-2012), so as to obtain the precipitation classification standards of different time scales.

4. The equal ranking rate method according to claim 1, characterized in that the ranking rates of different snowfall classification standards are obtained according to the 12-hour snowfall classification standard of the national standard (GB / T 28592-2012), so as to obtain the snowfall classification standards of different time scales.

5. The equal ranking rate method according to claim 1, characterized in that the ranking rates of different snowfall classification standards are obtained according to the 24-hour snowfall classification standard of the national standard (GB / T 28592-2012), so as to obtain the snowfall classification standards of different time scales.

Images