Meteorological precipitation observation data health degree identification method

Abstract

The invention provides a meteorological precipitation observation data health degree identification method. The meteorological precipitation observation data health degree identification method comprises the following steps: S1, obtaining an hourly precipitation data quality control code; s2, calculating characteristic parameters; s3, calculating rainfall data sub-health indexes; s4, calculating the overall health degree of the rainfall data; through the method, the health degree of the observation data can be accurately and objectively judged, so that a powerful guarantee is provided for the accuracy and reliability of meteorological precipitation data.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological observation, and in particular to a method for judging the health of meteorological precipitation observation data. Background Art

[0002] Data quality control plays a vital role in the high-quality development of meteorology. Through strict data quality control, the accuracy of meteorological data can be ensured, and errors caused by instrument failure or data entry errors can be avoided. High-quality meteorological data makes weather forecast services more accurate, provides the public with reliable weather information, and enhances public trust and satisfaction. Guizhou is located near the Hengduan Mountains and is also one of the areas with the most serious meteorological disasters and meteorological derivative disasters (flash floods, landslides, mudslides, etc.). These meteorological disasters and derivative disasters will block the main transportation lines, seriously damage power and communication facilities, damage crops, and freeze livestock to death, directly causing great losses to the province's economy, society, and people's lives and property safety, and have a great impact and difficulty on people's lives and production. The accuracy and reliability of precipitation data are crucial to weather forecasting. The precipitation data quality control methods currently used in business systems mainly include missing value checks, climate limit value checks, range value checks, time consistency checks, and spatial consistency checks. There is a lack of quality control methods for multi-source data. The 1km multi-source data fusion product produced by Guizhou Province mainly includes observation data, radar, satellite and other observation data. The present invention provides a method for identifying the health of meteorological precipitation observation data using multi-source fusion products. Summary of the invention

[0003] The purpose of the present invention is to provide a method for determining the health of meteorological precipitation observation data to solve the problems existing in the above-mentioned prior art.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] The present invention provides a method for judging the health of meteorological precipitation observation data, comprising the following steps:

[0006] S1. Obtain the hourly precipitation data quality control code;

[0007] S2. Calculate characteristic parameters;

[0008] S3. Calculate the health index of precipitation data;

[0009] S4. Calculate the overall health of precipitation data.

[0010] Preferably, step S1 is specifically as follows: first, determine whether to enter the next identification process based on the precipitation quality control code. If the quality control code is 2 or 8, the health of the precipitation data at that time is 0; if the precipitation quality control code is 0, 1, 3, or 4, continue with the subsequent steps.

[0011] Preferably, step S2 is specifically as follows: using 30 days of hourly historical precipitation observation data from 3000 stations and 1km precipitation real-time grid products, calculate the total deviation, relative deviation and correlation coefficient of all stations and the grid products, calculate and obtain 3 times the standard deviation of the corresponding average values ​​of the total deviation, relative deviation and correlation coefficient of the evaluation indicators of all observation stations, so as to obtain the maximum and minimum values ​​corresponding to each evaluation indicator, that is, the threshold value of the evaluation indicator.

[0012] Preferably, the characteristic parameters include: deviation, absolute deviation, rain difference rate, clear difference rate, and correlation coefficient, wherein:

[0013] The calculation formula for the deviation is:

[0014]

[0015] The absolute deviation is calculated as:

[0016]

[0017] The calculation formula of rainfall difference rate is:

[0018]

[0019] The calculation formula of the clearness difference ratio is:

[0020]

[0021] The formula for calculating the correlation coefficient is:

[0022]

[0023]

[0024] Where n is the total number of samples in the testing period of the testing site, O i is the precipitation observation value at the detection station, B i is the grid point estimate, nP and nS are the number of samples with and without precipitation in the station observation value and without and without precipitation in the grid point estimate value during the detection period.

[0025] Preferably, step S3 is specifically as follows: the health of the actual hourly precipitation is calculated using the evaluation index threshold and the standard value. When the actual value exceeds the evaluation index threshold range, the health of the precipitation data is 0; when the actual value is the same as the standard value, the health of the precipitation data is 1; when the actual value is within the evaluation index threshold range, the normalized data value (0, 1) is calculated based on the distance from the standard value as the health of the evaluation index of the precipitation data.

[0026] Preferably, the calculation method of the health degree is:

[0027] 1) Calculate the health of the observed data c1, and divide the total precipitation into different levels, namely: 0mm, 0.1-10mm, 10-50mm, 50-100mm, 100-150mm, and greater than 150mm. Select different evaluation indicators according to different levels of precipitation, and find the standard deviation of the corresponding deviation under the total precipitation classification. Assuming that the deviation is subject to normal distribution, find the minimum value s corresponding to the three times the standard deviation min and the maximum value s max Calculate the total precipitation health index c1;

[0028] when but

[0029] when but

[0030] When s>s max ,s min When , c1=0;

[0031] 2) Calculate the health of the observation data c2 and c3, and calculate the standard deviation of the rainfall difference rate during this period through historical data And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ;

[0032] Let s be the rainfall difference rate and calculate the sub-health degree c2;

[0033] when but

[0034] when but

[0035] When s>s max ,s min When , c2=0;

[0036] Let s be the clear sky difference rate, and calculate the sub-health degree c3;

[0037] ​​when but

[0038] when but

[0039] When s>s max ,s min When , c3=0;

[0040] 3) Calculate the health of the observed data c4, and calculate the standard deviation of the correlation coefficient during this period through historical data And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ;

[0041] Let s be the correlation coefficient and calculate the score health c4;

[0042] when but

[0043] when but

[0044] When s>s max ,s min , then c4=0.

[0045] Preferably, step S4 is specifically: calculating the overall health Fit of the observed data, which is mainly divided into three cases:

[0046] 1) When the actual value exceeds the threshold range of the evaluation index, the precipitation data health Fit = 0;

[0047] 2) When the total precipitation of the observed data is 0 mm, the sub-health levels c1, c2, and c3 are selected, and a certain weight is given by the tomography method for comprehensive calculation to obtain Fit = 0.7*c1+0.15*c2+0.15*c3;

[0048] 3) When the total precipitation in the observed data is greater than 0 mm, the c1, c2, c3, and c4 sub-health indicators are selected for comprehensive calculation, and the tomography method is used to assign weights to each indicator, Fit = 0.28*c1+0.12*c2+0.12*c3+0.48*c4.

[0049] Compared with the prior art, the present invention has achieved the following beneficial technical effects:

[0050] ​​The present invention provides a method for judging the health of meteorological precipitation observation data, comprising the following steps: S1. obtaining a quality control code for hourly precipitation data; S2. calculating characteristic parameters; S3. calculating a health index of precipitation data; S4. calculating the overall health of precipitation data; the above method can accurately and objectively judge the health of observation data, thereby providing a strong guarantee for the accuracy and reliability of meteorological precipitation data. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0052] Figure 1 A flowchart of a method for identifying the health of meteorological precipitation observation data provided by the present invention. DETAILED DESCRIPTION

[0053] 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.

[0054] The purpose of the present invention is to provide a method for determining the health of meteorological precipitation observation data to solve the problems existing in the prior art.

[0055] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0056] Embodiment 1:

[0057] This embodiment provides a method for judging the health of meteorological precipitation observation data. Figure 1 As shown, the following steps are included:

[0058] S1. Obtain the hourly precipitation data quality control code; first determine whether to enter the next identification process based on the precipitation quality control code. If the quality control code is 2 or 8, the health of the precipitation data at that time is 0; if the precipitation quality control code is 0, 1, 3, or 4, proceed to the subsequent steps;

[0059] S2. Calculate characteristic parameters; use 30 days of hourly historical precipitation observation data from 3000 stations and 1km precipitation real-time grid products to calculate the total deviation, relative deviation and correlation coefficient of all stations and grid products, and calculate 3 times the standard deviation of the corresponding average values ​​of the total deviation, relative deviation and correlation coefficient of all observation stations, so as to obtain the maximum and minimum values ​​corresponding to each evaluation indicator, that is, the threshold value of the evaluation indicator; among them, a total of 4019 stations in Guizhou Province (including national ground stations, first-class conventional stations, second-class conventional stations, third-class conventional stations, and fourth-class conventional stations) were selected. The hourly precipitation data of mm was read through the data service interface of Guizhou Meteorological Big Data Cloud Platform. The data has undergone initial quality control such as missing value check, climate limit value check, range value check, time consistency check and spatial consistency check;

[0060] "Guizhou Province Multi-source Fusion Real-time Analysis Hourly Precipitation Product (resolution: 1km, ground)" is a multi-source precipitation fusion analysis product with high temporal and spatial resolution in our province. It is suitable for refined meteorological grid real-time analysis, model forecast verification, small and medium-scale extreme precipitation event monitoring, hydrological models, etc. It can provide efficient, high-quality and high-resolution precipitation products for business applications and research such as weather forecasting, disaster prevention and mitigation. The data set is developed using observation data and numerical model products from multiple sources, including ground, radar, satellite, etc., and adopts key technologies such as deviation correction and fusion analysis. Data products are generated in real time and stored on the Guizhou Meteorological Big Data Cloud Platform;

[0061] The characteristic parameters include: deviation, absolute deviation, rain difference rate, clear difference rate, and correlation coefficient, among which:

[0062] The calculation formula for the deviation is:

[0063]

[0064] The absolute deviation is calculated as:

[0065]

[0066] The calculation formula of rainfall difference rate is:

[0067]

[0068] The calculation formula of the clearness difference ratio is:

[0069]

[0070] The formula for calculating the correlation coefficient is:

[0071]

[0072]

[0073] Where n is the total number of samples in the testing period of the testing site, O i is the precipitation observation value at the detection station, B i is the grid point estimate, nP and nS are the number of samples with or without precipitation in the station observation value but without or with precipitation in the grid point estimate value during the detection period;

[0074] S3. Calculate the health index of precipitation data;

[0075] The health of the actual hourly precipitation is calculated using the evaluation index threshold and standard value. When the actual value exceeds the evaluation index threshold range, the health of the precipitation data is 0; when the actual value is the same as the standard value, the health of the precipitation data is 1; when the actual value is within the evaluation index threshold range, the normalized data value (0, 1) is calculated based on the distance from the standard value as the sub-health of the evaluation index of the precipitation data; the calculation method of the sub-health is:

[0076] 1) Calculate the health of the observed data c1, and divide the total precipitation into different levels, namely: 0mm, 0.1-10mm, 10-50mm, 50-100mm, 100-150mm, and greater than 150mm. Select different evaluation indicators (deviation or absolute deviation) according to different levels of precipitation, and find the standard deviation of the deviation corresponding to each classification of total precipitation. Assuming that the deviation obeys a normal distribution, find the minimum value s corresponding to the three times the standard deviation. min and the maximum value s max Calculate the total precipitation health index c1;

[0077] when but

[0078] when but

[0079] When s>s max ,s min When , c1=0;

[0080] 2) Calculate the health of observation data c2 and c3, and calculate the standard deviation of the rain difference rate (or clear difference rate) during this period through historical data (grid products and observation products) And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ;

[0081] Let s be the rainfall difference rate and calculate the sub-health degree c2;

[0082] when but ​

[0083] when but

[0084] When s>s max ,s min When , c2=0;

[0085] Let s be the clear sky difference rate, and calculate the sub-health degree c3;

[0086] when but

[0087] when but

[0088] When s>s max ,s min When , c3=0;

[0089] 3) Calculate the health of the observation data c4, and calculate the standard deviation of the correlation coefficient during this period through historical data (grid product and observation product) And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ;

[0090] Let s be the correlation coefficient and calculate the score health c4;

[0091] when but

[0092] when but

[0093] When s>s max ,s min When , c4=0;

[0094] S4. Calculate the overall health of precipitation data and the overall health of observation data Fit, which is mainly divided into three cases:

[0095] 1) When the actual value exceeds the threshold range of the evaluation index, the precipitation data health Fit = 0;

[0096] 2) When the total precipitation of the observed data is 0 mm, the sub-health levels c1, c2, and c3 are selected, and a certain weight is given by the tomography method for comprehensive calculation to obtain Fit = 0.7*c1+0.15*c2+0.15*c3;

[0097] ​​​3) When the total precipitation in the observed data is greater than 0 mm, the c1, c2, c3, and c4 sub-health indicators are selected for comprehensive calculation, and the tomography method is used to assign weights to each indicator, Fit = 0.28*c1+0.12*c2+0.12*c3+0.48*c4.

[0098] The present invention uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for judging the health of meteorological precipitation observation data, characterized in that: The following steps are involved: S1. Obtain the hourly precipitation data quality control code; S2. Calculate characteristic parameters; S3. Calculate the health index of precipitation data; S4. Calculate the overall health of precipitation data.

2. The method for determining the health of meteorological precipitation observation data according to claim 1, characterized in that: Step S1 is specifically as follows: first, determine whether to enter the next identification process based on the precipitation quality control code. If the quality control code is 2 or 8, the health of the precipitation data at that time is 0; if the precipitation quality control code is 0, 1, 3, or 4, continue with the subsequent steps.

3. The method for determining the health of meteorological precipitation observation data according to claim 1, characterized in that: Step S2 is specifically as follows: using 30 days of hourly historical precipitation observation data from 3000 stations and 1km precipitation real-time grid products, calculate the total deviation, relative deviation and correlation coefficient of all stations and grid products, and calculate 3 times the standard deviation of the corresponding average value of the total deviation, relative deviation and correlation coefficient of the evaluation indicators of all observation stations, so as to obtain the maximum and minimum values ​​corresponding to each evaluation indicator, that is, the threshold of the evaluation indicator.

4. The method for determining the health of meteorological precipitation observation data according to claim 3 is characterized in that: The characteristic parameters include: deviation, absolute deviation, rain difference rate, clear difference rate, and correlation coefficient, among which: The calculation formula for the deviation is: The absolute deviation is calculated as: The calculation formula of rainfall difference rate is: The calculation formula of the clearness difference ratio is: The calculation formula of the correlation coefficient is: Where n is the total number of samples in the testing period of the testing site, O i is the precipitation observation value at the detection station, B i is the grid point estimate, nP and nS are the number of samples with and without precipitation in the station observation value and without and without precipitation in the grid point estimate value during the detection period.

5. The method for determining the health of meteorological precipitation observation data according to claim 3 is characterized by: Step S3 is specifically as follows: the health of the actual hourly precipitation is calculated using the evaluation index threshold and standard value. When the actual value exceeds the evaluation index threshold range, the health of the precipitation data is 0; when the actual value is the same as the standard value, the health of the precipitation data is 1; when the actual value is within the evaluation index threshold range, the normalized data value (0, 1) is calculated based on the distance from the standard value as the health of the evaluation index of the precipitation data.

6. The method for determining the health of meteorological precipitation observation data according to claim 5, characterized in that: The calculation method of the health degree is: 1) Calculate the health of the observed data c1, and divide the total precipitation into different levels, namely: 0mm, 0.1-10mm, 10-50mm, 50-100mm, 100-150mm, and greater than 150mm. Select different evaluation indicators according to different levels of precipitation, and find the standard deviation of the corresponding deviation under the total precipitation classification. Assuming that the deviation is subject to normal distribution, find the minimum value s corresponding to the three times the standard deviation min and the maximum value s max Calculate the total precipitation health index c1; when but when but When s>s max ,s min When , c1=0;​ 2) Calculate the health of the observation data c2 and c3, and calculate the standard deviation of the rainfall difference rate during this period through historical data And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ; Let s be the rainfall difference rate and calculate the sub-health degree c2; when but when but When s>s max ,s min When , c2=0;​ Let s be the clear sky difference rate, and calculate the sub-health degree c3; when but when but When s>s max ,s min When , c3=0;​ 3) Calculate the health of the observed data c4, and calculate the standard deviation of the correlation coefficient during this period through historical data And find the minimum value s of parameter s with three times the standard deviation min and the maximum value s max ; Let s be the correlation coefficient and calculate the score health c4; when but when but When s>s max ,s min , then c4=0.​ 7. The method for determining the health of meteorological precipitation observation data according to claim 6 is characterized by: Step S4 is specifically: calculating the overall health of the observed data Fit, which is mainly divided into three cases: 1) When the actual value exceeds the threshold range of the evaluation index, the precipitation data health Fit = 0; 2) When the total precipitation of the observed data is 0 mm, the sub-health levels c1, c2, and c3 are selected, and a certain weight is given by the tomography method for comprehensive calculation to obtain Fit = 0.7*c1+0.15*c2+0.15*c3; 3) When the total precipitation in the observed data is greater than 0 mm, the c1, c2, c3, and c4 sub-health indicators are selected for comprehensive calculation, and the tomography method is used to assign weights to each indicator, Fit = 0.28*c1+0.12*c2+0.12*c3+0.48*c4.

Images