Method for establishing a severe convective potential warning method

Abstract

The present invention discloses a severe convective potential warning method in the field of weather forecasting, comprising the following steps: parsing the source data of FY-4A and FY4B satellite products obtained by the latest algorithm, reading and outputting the data using the VC sharp programming language, and outputting the FY-4 data and image products of the severe convective process; analyzing the key atmospheric parameters for severe convective potential forecasting based on the ground and upper-air conventional meteorological sounding data and the model zero-field data of severe convective weather, and analyzing the weather characteristics of the product data after FY-4 parsing before the occurrence of severe convective weather; refining the severe convective warning and potential forecasting indicators, and further proposing a method for complementary warning of severe convective potential; conducting forecast verification on the obtained forecasting indicators and methods during the severe convective weather process, and the verification results show that the accuracy rate of the complementary warning method for severe convective potential proposed by the present invention is as high as over 60%, and the warning time in advance can reach 0.5 to 7 hours.

Description

Technical Field

[0001] The present invention belongs to the field of weather forecasting, and specifically relates to a method for establishing a severe convective potential warning method. Background Art

[0002] The plateau has complex terrain with large undulations. Under the action of terrain lifting and solar radiation heating, the thermal and water vapor distributions of the underlying surface are uneven. Local heat convection is likely to occur in the afternoon in summer, and severe convective weather, including short-term heavy precipitation, hail, thunderstorm gales, etc., is extremely likely to occur with the assistance of weather systems, and thunderstorm gales rarely appear in patches. The geological environment in the plateau area is fragile, and the phenomenon of warming and humidification has been severe in recent years. Therefore, there are many severe convective disasters, especially heavy precipitation disasters, and severe convection is characterized by strong locality and short duration. On the premise that meteorological stations in the plateau area are relatively sparse and numerical weather prediction models lack mature assimilation techniques for the complex underlying surface in the plateau, there is currently a lack of a set of mature severe weather potential prediction indicators and technical methods. It is difficult to rely solely on the model prediction of severe convective weather, and it is extremely easy to miss small and medium-scale severe convective weather, which restricts the further development of the objective prediction and warning services for severe weather in the plateau.

[0003] The new generation of meteorological satellites FY-4A (still in orbit) and FY-4B (abbreviation: "FY-4") developed in China have the characteristics of wide detection range and high spatio-temporal resolution, and are effective tools for monitoring and warning severe convective weather. The FY-4 satellite data has been officially "stable" in operation in the national meteorological service departments since 2018 and 2023 respectively. Its product data is of great significance for the prediction and warning of severe convective weather, especially the instantaneous quantitative precipitation estimation QPE, cloud type CLT, cloud top temperature CTT, cloud top height CTH, convective initiation CIX, tropopause folding TFP products of the AGRI instrument of the FY-4 satellite, the atmospheric temperature and humidity profile AVP product of the GIIRS instrument, the lightning detection LMIE product of the LMI instrument, etc. However, up to now, its meteorological business applications are very few, and its application potential has not been well developed.

[0004] Therefore, in order to have a set of severe convective weather potential prediction index systems for the plateau, and to enable the FY-4 satellite to fully play its role in monitoring and warning severe weather in the plateau area, improve the core support ability of severe weather forecasting services in the plateau area, and further improve the disaster prevention and mitigation capabilities of meteorological departments and flood control and drought relief command departments in the plateau area, etc., it is necessary to combine conventional meteorological data and numerical model prediction data with the product data of FY-4, and conduct in-depth research on the technical indicators and methods for severe weather potential prediction in the plateau area. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for establishing a severe convective potential warning method to solve the problems described in the background art.

[0006] A method for establishing a severe convective potential early warning method comprises the following steps:

[0007] Source data quality control: Compare the actual severe convective weather conditions, detect and remove outliers in conventional meteorological detection data and model zero-field data, and use existing satellite calibration and positioning technology to pre-process the FY-4 satellite source data;

[0008] Parsing source data and output data format: Parsing FY-4 product source data obtained by the latest algorithm, using VC sharp programming language to read and output data, and outputting FY-4 data and image products of severe convection processes;

[0009] Severe convection forecast method and index extraction: Based on the actual severe convective weather and conventional meteorological detection data as well as model zero-field data, analyze the key atmospheric parameters for severe convection potential forecast, analyze the weather characteristics of FY-4 product data before severe convective weather occurs, and extract severe convection warning and potential forecast indicators;

[0010] Severe convective forecast verification: The obtained forecast methods and forecast indicators are tested in severe convective weather processes to verify the forecast timeliness and accuracy of FY-4A product data in severe convective weather forecasts, as well as the scope of data application.

[0011] Furthermore, the source data of FY-4 products include tropopause fold TFP, atmospheric temperature profile AVP, convective initiation CIX, 1-minute lightning event LMIE, instantaneous quantitative precipitation estimate QPE, cloud type CLT, cloud top temperature CTT and cloud top height CTH; conventional meteorological detection source data include processed micaps format T_logP sounding data, wind field and temperature field and dew point field data from the ground to high altitude; severe convection real-time data include automatic weather station monitoring data and manually recorded real-time data; model zero field data uses convective effective potential energy CAPE data.

[0012] Furthermore, VC sharp programming language and micaps system are used to read and output data, and to extract forecast indicators and forecast methods, including:

[0013] The LMIE within 15 minutes is stacked and processed into the cumulative lightning monitoring data LMIEX, and the four FY-4 product data of TFP, CIX, LMIEX and CLT are generated into micaps type 3 data format;

[0014] Generate micaps Class 4 data format for the four FY-4 product data, CTH, CTT, QPE and TFP (including TROPPV, TROPH, Tropo_P_depth and Tropo_Z_depth);

[0015] Generate micaps Class 5 format data from AVP product data;

[0016] Generate shp format data from TFP, CIX, and LMIEX;

[0017] Generate tif format data from CTT and QPE;

[0018] Import the above output data of FY-4 into the micaps system for analyzing the performance characteristics of severe convective data and the warning indicators of severe convection;

[0019] Vertical wind shear: According to the wind field data from the ground to the upper air, calculate the vertical wind shear vector values VS 0-3 and VS 0-6 ;

[0020] Sounding correction: Compare the occurrence location, time, and inversion situation of the severe convection actual situation, select a sounding station within the same nature cloud cluster as the severe convection location, with a spatial distance not exceeding 150 km, and the sounding time before the severe convection time. Import the T_logP sounding data into the micaps system to display the T_logP image of the selected sounding station. If there is an inversion within 1 hour before the occurrence of severe convection, use the micaps system function to correct the T_logP image at the top of the inversion layer. If there is no inversion, use the actual ground temperature and dew point data closest to the severe convection location and time before the occurrence of severe convection to correct the T_logP image;

[0021] Sounding structure: Output the sounding structure identifier according to the corrected T_logP image;

[0022] CAPE value: Calculate the CAPE estimated value according to the corrected T_logP data, compare the CAPE value of the model zero field, and take the maximum value of the two;

[0023] Temperature difference T 400-600 in the middle and lower layers: Output the temperature difference from 600 hPa to 400 hPa at the severe convection occurrence location according to the corrected T_logP data;

[0024] Dry layer thickness △H dry and wet layer thickness △H wet : Output the dry layer thickness and wet layer thickness at the severe convection occurrence location according to the corrected T_logP data;

[0025] 0℃ layer height H0: Output the 0℃ layer height on the day of the severe convection occurrence location according to the corrected T_logP data;

[0026] Dynamic lifting conditions: The wind field and temperature field data from the ground to the high altitude are transferred into the MICAPS system to determine the middle and low-level wind field and temperature field where the severe convection occurs, and output the signs of whether there is wind field shear, convergence and fluctuation, and whether there is a cold temperature trough.

[0027] The above conventional detection and model zero field and its output data are transferred into the micaps system to analyze the severe convection potential forecast indicators;

[0028] Summarize all analysis results, refine severe convection warning indicators, and condense severe convection warning methods.

[0029] Furthermore, the FY-4 product data are from the National Satellite Meteorological Center and its website, and the actual weather conditions and conventional meteorological detection data as well as the model zero-field data are from the local meteorological bureau.

[0030] Furthermore, the update frequency of the FY-4 tropopause folded TFP data is 60 minutes, the update frequency of the atmospheric temperature profile AVP and the convection incipient CIX data is 15 minutes, the update frequency of the 1-minute lightning event LMIE data is 1 minute, the update frequency of the instantaneous quantitative precipitation estimation QPE data, the cumulative lightning events LMIEX, the cloud type CLT, the cloud top temperature CTT, and the cloud top height CTH data are all 15 minutes; the update frequency of conventional detection and model zero-field meteorological data is 12 hours.

[0031] Furthermore, the atmospheric temperature profile AVP is specifically obtained by inverting the regional atmospheric temperature profile product from the GIIRS channel matching dataset of the FY-4 satellite.

[0032] Furthermore, the key atmospheric parameters for severe convective potential forecast are analyzed, including sounding structure, unstable energy, wind shear, meteorological element field and circulation situation;

[0033] The data performance after FY-4 product analysis is analyzed before severe convective weather occurs, where severe convective weather includes hail, short-term heavy rainfall and thunderstorm gale.

[0034] The above scheme has the following beneficial effects:

[0035] In the present invention, according to the actual situation of severe convection, the severe convection potential forecasting conditions and indicators of conventional meteorological sounding data (upper air, surface) and numerical model zero-field data are analyzed, and the severe convection early warning conditions and indicators after the analysis of FY-4 products are analyzed; a method of combining conventional meteorological sounding and model data with FY-4 analysis products complementarily for severe convection early warning is proposed, and the early warning indicators and methods are summarized and refined. Compared with the prior art, the accuracy rate of the early warning method of the present invention is high. The accuracy rates of separately conducting severe convection potential forecasting, severe convection early warning of FY-4 analysis products, or conducting complementary severe convection potential early warning all reach or exceed 60%. Among them, the complementary early warning method is superior to the forecasting and early warning of the former two separately in terms of accuracy and timeliness.

[0036] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flowchart of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0038] Figure 2 It is a schematic diagram of an example of the micaps type 3 format of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0039] Figure 3 It is a schematic diagram of an example of the micaps type 4 format of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0040] Figure 4 It is a schematic diagram of an example of the micaps type 5 format of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0041] Figure 5 It is an analysis diagram of the sounding structure and severe convection potential of the hail process on June 13 of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0042] Figure 6 It is the CIX recognition result at 15:38 on June 13 in Banma County of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0043] Figure 7 It is the CLT monitoring result at 16:38 on June 13 in Banma County of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0044] Figure 8 It is the CTT monitoring result at 17:53 on June 13 in Zadoi County of an embodiment of the method for establishing the severe convection potential early warning method of the present invention;

[0045] Figure 9Monitoring results of Dari County at 16:15 CTT on June 30th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0046] Figure 10 Monitoring results of Banma County at 16:38 CTH on June 13th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0047] Figure 11 Monitoring results of Zadoi County at 17:30 CTH on June 13th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0048] Figure 12 Monitoring results of Guinan County at 18:53 QPE on August 5th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0049] Figure 13 Monitoring results of Henan County at 19:38 QPE on June 15th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0050] Figure 14 Monitoring results of the eastern part of Qinghai Lake at 15:00 TFP on August 24th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0051] Figure 15 Monitoring results of Huangyuan County and Huangzhong County at 16:00 TFP on August 24th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0052] Figure 16 Monitoring results of Zhidoi Station at 14:30 AVP on June 13th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0053] Figure 17 Monitoring results of Banma Station at 16:30 AVP on June 13th for the example of the establishment method of the severe convective potential warning method of the present invention;

[0054] Figure 18 Monitoring results of Jianzha County at 20:30 - 20:45 LMIEX on August 21st for the example of the establishment method of the severe convective potential warning method of the present invention. Detailed implementation manners

[0055] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0056] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "longitudinal", "transverse", "vertical", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention.

[0057] In the description of the present invention, unless otherwise specified and defined, it should be noted that the terms "installation", "connection", and "coupling" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the communication inside two elements. It can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0058] The following is a further detailed description through specific embodiments:

[0059] The embodiment is as shown in the attached Figure 1 The establishment method of the severe convective potential warning method includes the following steps:

[0060] S1. Source data quality control: Compare the severe convective weather facts, perform data outlier detection and elimination on the conventional meteorological sounding data and model zero-field data, and preprocess the source data of FY-4A and FY4B (hereinafter referred to as FY-4) satellite products using the existing satellite calibration and positioning technology.

[0061] S2. Analyze the source data and output data format: Analyze the FY-4 product source data obtained by the latest algorithm, use the VCsharp programming language and the micaps system to read and output the data, and output the FY-4 data and image products of the severe convective process.

[0062] Among them, the FY-4 product source data includes tropopause fold (TFP), atmospheric temperature profile (AVP), convective initiation (CIX) (in the case where CIX data is missing for FY-4B, the CIX data of FY-4A is used for supplementation), 1-minute lightning event (LMIE), instantaneous quantitative precipitation estimation (QPE), cloud type (CLT), cloud top temperature (CTT), and cloud top height (CTH); the conventional meteorological sounding source data includes the processed micaps format T_logP sounding data, wind field, temperature field, and dew point field data from the ground to high altitude; the severe convective fact data includes the automatic weather station monitoring data and the manually recorded fact data; the model zero-field data uses the convective available potential energy (CAPE) data.

[0063] Specifically, the TFP product of the AGRI instrument on FY-4 satellite: It reflects the process of the air with high stratospheric potential vorticity squeezing into the troposphere, which often causes cyclogenesis, strong convection, and clear-air turbulence. It includes a dataset of 7 parts, all of which are 2D float-type data with 2748 rows × 2748 columns for the full disk: lat (latitude), lon (longitude), TROPPV (tropopause pressure, unit: hPa), TROPH (tropopause height, unit: m), TFTP (tropopause fold, with 127 as the fill value), Tropo_P_depth (thickness of the tropopause fold, unit: hPa), Tropo_Z_depth (thickness of the tropopause fold, unit: km). The update frequency of this data used in this embodiment is 60 minutes.

[0064] The AVP product of the GIIRS instrument on FY-4 satellite: Preferably, the dwell point (sub-region scan) product is used. Specifically, it is a regional atmospheric temperature profile product retrieved from the GIIRS channel matching dataset, with a resolution of 16 km. This type of data generally contains a total of 59 dwell point numbers every 15 minutes, that is, one dwell point data is completed every about 15 seconds. The commonly used data blocks are AT_prof (atmospheric temperature at 101 layer heights × 128 horizontal positions, unit: K), AT_prof_Qflag (data quality flag for atmospheric temperature data at 101 layer heights × 128 horizontal positions), Pressure (pressure values of each layer at 101 layer heights, unit: hPa), LW_Latitude (latitude at 128 horizontal positions), LW_Longitude (longitude at 128 horizontal positions).

[0065] The CIX product of the AGRI instrument on FY-4 satellite: It has a resolution of 4 km and is commonly used for the identification of the initiation and development of convection in nowcasting. The commonly used datasets are CI (convection initiation) and Mature (mature convection). The update frequency of this data used in this embodiment is 15 minutes.

[0066] The LMIE product of the LMI instrument on FY-4 satellite, that is, the L2-level 1-minute quantitative product, has a resolution of 7.8 km and an update frequency of 1 minute. It is commonly used for early warning of severe convective weather processes, identification of convection initiation, and research on atmospheric electric fields, etc. The commonly used data blocks are DQF (data quality flag for lightning events), ER (radiation intensity), LAT (latitude), LON (longitude).

[0067] The QPE product of the AGRI instrument on FY-4 satellite, that is, the quantitative precipitation estimation product, has a resolution of 4 km, and the data update frequency ranges from several minutes to 60 minutes. The data update frequency used in this study is 15 minutes. The commonly used data blocks are DQF (data quality flag), Precipitation (instantaneous precipitation).

[0068] The CLT (Cloud Type) product of the AGRI instrument on FY-4 satellite has a resolution of 4 km, and the data update frequency ranges from several minutes to 60 minutes. The data update frequency used in this study is 15 minutes. The common data block is a two-dimensional data CLT of 1092 rows × 2748 columns, with a value of 0 representing clear sky, 2 representing precipitation, 3 representing supercooled water, 4 representing mixed type, 5 representing ice and snow, 6 representing stratus cloud, and 7 representing overlapping type.

[0069] The CTT (Cloud Top Temperature) product of the AGRI instrument on FY-4 satellite uses data detected by multiple infrared channels to invert and synthesize cloud top temperature data (unit: K), with a resolution of 4 km. The data update frequency ranges from several minutes to 60 minutes. The data update frequency used in this study is still 15 minutes. The common data block is a two-dimensional data CTT of 1092 rows × 2748 columns.

[0070] The CTH (Cloud Top Height) product of the AGRI instrument on FY-4 satellite uses data detected by multiple infrared channels to invert and synthesize cloud top height data (unit: m), with a resolution of 4 km. The data update frequency ranges from several minutes to 60 minutes. The data update frequency used in this study is still 15 minutes. The common data block is a two-dimensional data CTH of 1092 rows × 2748 columns.

[0071] Among them, the VC sharp programming language and the micaps system are used to read and output data, specifically including:

[0072] The LMIE within 15 minutes is superimposed and processed into cumulative lightning monitoring data LMIEX, and the data of four FY-4 products, namely TFP, CIX, LMIEX, and CLT, are generated into the micaps type 3 data format;

[0073] The data of four FY-4 products, namely CTH, CTT, QPE, and TFP (including TROPPV, TROPH, Tropo_P_depth, and Tropo_Z_depth), are generated into the micaps type 4 data format;

[0074] The AVP product data is generated into the micaps type 5 format;

[0075] The data of TFP, CIX, and LMIEX are generated into shp format data;

[0076] The data of CTT and QPE are generated into tif format data;

[0077] Vertical wind shear: According to the wind field data from the ground to the high altitude, the vertical wind shear vector values VS 0-3 and VS 0-6 ;

[0078] Sounding correction: Compare the location and time of the actual strong convection and the inversion situation, select the sounding station that is in the same cloud cluster as the strong convection location, with a spatial distance of no more than 150km and a sounding time before the strong convection time, and transfer the T_logP sounding data into the micaps system to display the T_logP image of the selected sounding station. If there is an inversion within 1 hour before the occurrence of strong convection, use the micaps system function to correct the T_logP image at the top of the inversion layer. If there is no inversion, use the actual ground temperature and dew point data closest to the location and time of the strong convection before the occurrence of strong convection to correct the T_logP image;

[0079] Sounding structure: Output the sounding structure identification according to the corrected T_logP image;

[0080] CAPE value: Calculate the CAPE estimate based on the corrected T_logP data, compare the CAPE value of the model zero field, and take the maximum value of the two;

[0081] Temperature difference between middle and lower layers T 400-600 :According to the corrected T_logP data, output the temperature difference from 600hPa to 400hPa where the severe convection occurs;

[0082] Dry layer thickness △H dry and wet layer thickness △H wet : According to the corrected T_logP data, output the dry layer thickness and wet layer thickness where severe convection occurs;

[0083] 0℃ layer height H0: According to the corrected T_logP data, the 0℃ layer height of the day when the severe convection occurred is output;

[0084] Dynamic lifting conditions: The wind field and temperature field data from the ground to the high altitude are transferred into the MICAPS system to determine the middle and low-level wind field and temperature field where the severe convection occurs, and output the signs of whether there is wind field shear, convergence and fluctuation, and whether there is a cold temperature trough.

[0085] Specifically, micaps type 3 format data: This type of format data can display the corresponding site value on the micaps system platform. This study generates Micaps type 3 data format from four FY-4A product data blocks, including tropopause folding (TFP), convective inception (CIX), lightning events (LMIEX), and cloud types (CLT), as shown in the attached figure. Figure 2 The data format is:

[0086] diamond 3 data description (string)

[0087] Year, month, day and time level

[0088] Number of contour lines (all integers)

[0089] Contour value 1 Contour value 2 Smoothing coefficient

[0090] Number of single-station plotting elements Total number of stations (both are integers)

[0091] Station number (long integer) Longitude Latitude Altitude (all are floating-point numbers) Station value 1 Station value 2 (both are strings).

[0092] MICAPS Category 4 format data:

[0093] This type of format data can display corresponding contour lines on the MICAPS system platform. In this study, data blocks of FY-4 products such as cloud top height (CTH), cloud top temperature (CTT), quantitative precipitation estimation (QPE), and tropopause fold (TFP), including TROPPV, TROPH, Tropo_P_depth, and Tropo_Z_depth, are used to generate MICAPS Category 4 data format, as shown in the appendix Figure 3 as follows. The data format is:

[0094] Diamond 4 data description (string)

[0095] Year Month Day Hour Forecast hour Forecast level (all are integers) Longitude grid interval Latitude grid interval Starting longitude Ending longitude Starting latitude Ending latitude (all are floating-point numbers) Number of grid points in the zonal direction Number of grid points in the meridional direction (both are integers) Contour interval Contour starting value Ending value Smoothing coefficient Bold line value (all are floating-point numbers. Changing the bold line value to -1 means plotting the contour and filling the map simultaneously, and changing it to -2 means only filling the map without plotting the contour)

[0096] Grid point data block: The data is arranged first by latitude and then by longitude (in a rectangular coordinate grid, it is first in the X direction and then in the Y direction), and all are floating-point numbers.

[0097] MICAPS Category 5 format data:

[0098] This type of format data can display corresponding profiles on the MICAPS platform. In this study, the atmospheric temperature profile (AVP) data of FY-4 products are used to generate the MICAPS Category 5 format, as shown in the appendix Figure 4 as follows. The data format is:

[0099] Diamond 5 data description (string)

[0100] Year Month Day Hour Total number of stations (all are integers)

[0101] Station number 1 Longitude Latitude Altitude Length of single-station content

[0102] Pressure Height Temperature Dew point Wind direction Wind speed at the first layer

[0103] Second layer: air pressure, altitude, temperature, dew point, wind direction, wind speed

[0104] Station number, longitude, latitude, altitude above sea level, length of single-station content

[0105] (The length of single-station content is the number of layers × 6; when there are missing values except for wind direction and wind speed, the entire layer is cancelled, and missing values of wind direction and wind speed are represented by 9999)

[0106] shp format data:

[0107] The shp file consists of an index file (*.shx), a dBASE table (*.dbf), and a main file (*.shp). The main file consists of a 100-byte file header and variable-length records. The index file mainly consists of records of offsets from the start of the main file header. The dBASE table is the attribute table. The output shp format data can be loaded in geographic information software to facilitate viewing data ranges that meet specific conditions, etc.

[0108] tif format data:

[0109] The tif data consists of an 8-byte file header and an image file directory composed of consecutive tags containing various image information. The output tif format data can be used to judge the grading range of data values according to image colors and to load map boundaries in geographic information software. In this study, CTT and QPE output micaps grid data and tif format data with a data resolution of 4 km.

[0110] S3. Severe convective weather forecasting methods and index extraction: Combining the actual situation of severe convective weather, the conventional sounding and model initial fields and their output data in S2 are imported into the micaps system to analyze the key atmospheric parameters and their characteristics for severe convective potential forecasting before the occurrence of severe convective weather; the FY-4 output data in S2 is imported into the micaps system to analyze the weather characteristics shown by the FY-4 data before the occurrence of severe convective weather; summarize and refine the severe convective potential forecasting and warning indicators and methods.

[0111] Among them, severe convective weather includes hail, short-term heavy precipitation, and thunderstorm winds; analyze the key atmospheric parameters for severe convective potential forecasting, including sounding structure, unstable energy, wind shear, meteorological element fields, and circulation patterns; analyze the weather characteristics shown by the FY-4 analysis product data before the occurrence of severe convective weather, including all FY-4 output product data in S2.

[0112] Among them, the FY-4 product data comes from the National Satellite Meteorological Center and its website, and the actual weather conditions, conventional meteorological sounding data, and model initial field data come from the local meteorological bureau. This invention mainly uses the severe convective weather process data in 2020 and 2021, and the time in the source data file names of FY-4 products is in Universal Time.

[0113] As attachedFigure 5 - Appendix Figure 18 As shown below, taking the severe convective potential warning in Qinghai region as an example:

[0114] Using the typical severe convective weather processes in 2020 (see Table 1), analyze their circulation patterns, key meteorological elements, sounding structures, etc., so as to obtain the severe convective potential forecasting indicators and methods for the mesoscale atmosphere.

[0115] Table 1 Information Table of Main Severe Convective Weather Processes in 2020

[0116]

[0117]

[0118] Comprehensively analyzing the above-mentioned severe convective potential and sounding analysis results, before the approach of hail and thunderstorm with strong winds, there is a certain dry layer in the middle and upper troposphere. The sounding diagram is mostly in the shape of an X, that is, there is a certain wet layer in the middle layer, and the atmosphere in the other layers is relatively dry. The vertical wind shear in the deep layer and the low layer generally approaches or reaches medium intensity (about 10 m / s and above). The dynamic lifting conditions are sufficient. Generally, there is a ground wind field convergence line or a low and middle layer wind shear, and the temperature difference between the low and middle layers is relatively large, with the temperature difference between the low and middle layers at least greater than 16 °C. Before the occurrence of hail weather, the CAPE value after sounding correction and the CAPE value of the model zero field are mostly greater than 600 J / kg, and for large hailstones, it mostly exceeds 1000 J / kg, and even exceeds 3000 J / kg. Compared with hail weather, the CAPE value of thunderstorm with strong winds is smaller or even has no energy, but the dry layer is deeper and there is a certain convective available potential energy of subsidence. For thunderstorm weather mainly with convective strong winds, the dry layer even reaches 200 hPa, and the near-surface atmosphere is very dry. Before the approach of short-term heavy precipitation, the atmospheric humidity conditions are better than those of hail and thunderstorm with strong winds. Generally, the wet layer is deep and at least reaches 200 hPa thick. The CAPE value is generally greater than 200 J / kg, and the area where CAPE is located on the sounding diagram is generally slender. The vertical wind shear in the deep layer can be large or small. In addition, the atmospheric conditions near the above-mentioned severe convective locations are quite the same, but no severe convective weather occurs, which is all related to the complex terrain and underlying surface in Qinghai. The terrain lift, ground heat, water vapor convergence, etc. at the severe convective occurrence location should be more conducive to the triggering or strengthening of severe convection.

[0119] Using the FY-4A product data (nc and hdf formats) corresponding to the main severe convective weather processes in 2020 (see Table 1), output the micaps format, shp and tif format data that are convenient for forecast analysis.

[0120] Within 1 hour before the approach of severe convective weather, the cloud type (CLT product) is mainly composed of ice-phase particles. The cloud type is most likely to be ice-phase cloud particles whether it is short-term heavy precipitation or hail weather. Especially before the approach of hail weather or thunderstorm weather phenomena, the cloud type is mainly ice-phase particles. For example, on June 14, there were hail and short-term heavy precipitation weather, on June 15, there were hail and short-term heavy precipitation weather, on August 5, there were short-term heavy precipitation and thunderstorm weather, on August 24, there were hail and thunderstorm gale weather, etc. However, before the approach of severe convection mainly characterized by convective gales, the cloud type can be a chaotic type mixed with multiple cloud types. For example, on July 2, there was a weather process mainly characterized by convective gales accompanied by several weak thunderstorms. CLT is closely related to the cloud top temperature. When the cloud top temperature CTT is greater than 245K, the cloud type CLT is mostly precipitation type.

[0121] Before the occurrence of most severe convective weather, the convective initiation (CIX) product can identify that there is mature convection covering the sky above the severe convective location. The CIX has a poor effect on identifying convection in local severe convective processes closely related to complex underlying surfaces. For example, in the hail on July 8, the short-term heavy precipitation on August 5, the short-term heavy precipitation weather on August 21, etc., no incipient or mature convection above the severe convective location was identified. Sometimes, the CIX may have missing data. For example, there is no CIX data during the severe convective periods on August 24 and August 26.

[0122] Before the approach of severe convection, the cloud top brightness temperature product (CTT) mainly shows a decrease in the cloud top brightness temperature. The low brightness temperature area is generally below 240K. Severe convection generally occurs in areas where the cloud top brightness temperature is below 255K, in the low brightness temperature center area or in areas with a large brightness temperature gradient where the brightness temperature difference between adjacent pixels is greater than 5K. Hail occurs more in areas with a large brightness temperature gradient. Thunderstorm gales generally occur at the edge of the low brightness temperature area, outside the area with a large brightness temperature gradient of the cloud top, and the cloud top brightness temperature is about 250K - 265K.

[0123] Before the approach of severe convective weather, the cloud top height product (CTH) shows a value reaching or exceeding 9 km, and can even reach 20 km. Before the occurrence of severe convection, the cloud top height can rise by 2 km to 5 km, or remain at a height of about more than 11 km. The change in the CTH height corresponds to the change in the cloud top brightness temperature of CTT. The location where the cloud top brightness temperature of CTT becomes lower corresponds to the rise in the cloud top height of CTH. If the convective cloud cluster changes slowly, the CTH height remains basically unchanged; if it changes rapidly, the CTH height changes rapidly. The CTH of the weather process with a mixture of various severe convective weather is relatively high, maintaining an average height of 15 km and above. For example, within 2 hours before the occurrence of severe convection accompanied by short-term heavy precipitation and lightning on August 5th, the CTH was 13 km to 18 km. Before the occurrence of short-term heavy precipitation, hail and lightning on June 30th, the cloud top height was 14 km to 20 km. Half an hour before the approach of hail, short-term heavy precipitation and severe thunderstorm winds on June 16th, the cloud top height remained at 17 km to 18 km. Before the occurrence of severe thunderstorm winds mainly characterized by convective winds on July 2nd, the cloud top height was 7 km to 9 km, which was lower than the tropopause height of other severe convective weather.

[0124] During the severe convective weather process in summer in Qinghai, there are few cases with clear characteristics of the tropopause folding product (TFP), that is, the number of times of the intrusion of high-potential vorticity air in the stratosphere into the troposphere reflected from FY-4 satellite data is small. There are only two processes with the phenomenon of tropopause folding in the above severe convective weather processes. For example, during the process of severe convective weather in many places in northeastern Qinghai on July 8th, there was a tropopause folding phenomenon existing for a long time (more than 12 hours) in the north of the northern provincial boundary of Qinghai. During the severe convective process on August 24th, there was a tropopause folding phenomenon above or near the severe convective location. The common feature of these two processes is that there were severe thunderstorm winds for a long time (the continuous occurrence time reached 3 hours and above), and the convective weather was intense, with hailstones with a diameter greater than 20 mm and precipitation with an hourly precipitation exceeding 25 mm occurring respectively. Therefore, strong convective weather will definitely occur at or near the location where tropopause folding appears in summer, and there is a high probability of severe thunderstorm winds.

[0125] The temperature profile product (AVP) shows the characteristics of a conditionally unstable atmospheric stratification before the occurrence of severe convection, and the AVP shows the characteristics of deep convection with a deep wet layer for the severe convective process on August 5th. Due to the limitation of the scanning method, the one-minute lightning monitoring (LMIE) product of FY-4 has data only within a certain dwell number and time range in the Qinghai area. Therefore, the available data of LMIE and LMIEX in the severe convective process in Qinghai are very limited, and currently, the indication of cloud flashes monitored by LMIE for severe convection is also limited.

[0126] In addition, by using conventional meteorological sounding and model initial field data and FY-4A analysis products, a comparative analysis of severe convective processes on July 8 and August 24 shows that the locations with tropopause folding are under the cold trough bottom of the guiding airflow layer for a long time (08:00 - 20:00). This indicates that after the high-potential vorticity air in the stratosphere intrudes into the troposphere, cold air will sink, leading to an increase in the temperature difference between the middle and lower layers, thus generating or enhancing convective weather.

[0127] Based on the above analysis, the forecasting and warning indicators and methods for severe convective weather within 3 hours before the occurrence of severe convection in the Qinghai Plateau can be summarized as shown in Table 2. The sounding structure is of the "X" type or approximately the "\\" type, that is, dry in the upper layer and wet in the lower layer or relatively wet throughout the layer. Most have a certain convective available potential energy (CAPE), with the lifting effect of synoptic-scale or terrain-induced mid-low-level shear lines or terrain convergence lines, significant temperature drops, or the temperature difference between the middle and lower layers generally greater than 16°C. Due to the complex terrain in Qinghai and the sparsity of meteorological stations, especially sounding stations, only sounding data at 08:00 and 20:00 are available. Therefore, it is possible that the conditions for severe convective potential in Table 2 are not met. Thus, it is acceptable that all the indicators for severe convective potential in Table 2 are satisfied simultaneously, or all the indicators of the FY-4 products are satisfied simultaneously.

[0128] The sounding structures of disaster-causing hail and thunderstorm gale weather are mostly of the X type. There must be a dry layer with a certain thickness in the middle and upper layers, and the thickness of the dry layer reaches or exceeds that of the wet layer. The requirement for water vapor is relatively low, but there must be a certain humidity in the middle or lower layers. The 0°C layer height on the day of hail weather is lower than 6.2 km, and the vertical wind shear in the middle or lower layers generally reaches or exceeds moderate wind shear intensity (i.e., at least ≥9 m / s). In a few cases, hail may also be generated under conditions such as terrain lifting and relatively small vertical wind shear over complex underlying surfaces, but the CAPE is generally greater than 1000 J / kg. The ground to near-surface atmosphere of thunderstorm gale weather, especially thunderstorm gales dominated by convective gales, is very dry. The dry layer is deeper than that of hail weather and generally reaches 250 hPa, and the thickness of the dry layer far exceeds that of the wet layer. And there is generally downdraft convective available potential energy. However, due to the sparsity of sounding stations on the plateau, sometimes the location of thunderstorm gales is not within the effective range of sounding stations. After sounding correction, the CAPE value of hail weather is generally greater than 600 J / kg, while that of thunderstorm gale weather is smaller or even shows no CAPE energy. When hail weather is accompanied by thunderstorm gales, the sounding and element characteristics of hail prevail.

[0129] The sounding of short-term heavy precipitation weather is most obvious in the form of a thick wet layer with a close temperature and dew point profile. The thickness of the wet layer reaches or exceeds the thickness of the dry layer. The CAPE value after sounding correction is generally greater than 200J / kg. The area where the CAPE is located on the sounding map is generally elongated, and the vertical wind shear can be small or large. The sounding shape of thunderstorms or thunderstorm gales accompanied by short-term heavy precipitation weather is still similar to the "\\" type. The thickness of the wet layer still reaches or exceeds 150hPa but is less than the thickness of the dry layer, and the middle and upper dry layers have a "V"-shaped opening.

[0130] Table 2. Plateau severe convection potential forecast and severe convection warning conditions of FY-4 analytical products

[0131]

[0132]

[0133] Within 1 hour before the occurrence of severe convection near the plateau, the temperature profile product AVP of FY-4A is a conditionally unstable profile between the dry and wet profiles. The nascent convection product CIX can generally identify convection in the nascent or mature convection stage, but both may lack data. Hail weather and short-term heavy precipitation weather are mainly severe convection of cold cloud nature. The cloud top brightness temperature product CTT of FY-4A product is generally not higher than 240K, but the convective clouds of short-term heavy precipitation weather are sometimes warmer (CTT greater than 240K). The cloud type CLT is mainly ice phase, followed by ice phase and mixed particles. Severe convection generally occurs in low brightness temperature areas or areas where the cloud top brightness temperature difference between adjacent pixels is greater than 5K. Short-term heavy precipitation mostly occurs in areas with CTT lower than 255K, while the CTT of hail production locations is higher. When the temperature profile of AVP is parallel to the wet adiabatic line, short-term heavy precipitation weather often occurs. The cloud top height product CTH for hail and short-term heavy rainfall is generally not less than 9km. The instantaneous precipitation product QPE for short-term heavy rainfall is generally greater than 10mm, but it is underestimated for areas with CTT higher than 240K. The estimated QPE value for hail weather is generally less than 6mm, but it is overestimated for areas with CTT lower than 240K. The convective clouds in thunderstorm or thunderstorm gale weather are warmer (CTT higher than 240K). Thunderstorm gale generally occurs near CTT higher than 250K and the brightness temperature difference of cloud tops of adjacent pixels is greater than 5K, and CTH is mostly lower than 9km. The CTH of mixed severe convective weather containing two or more severe convective phenomena is generally higher than 15km. Sometimes, the tropopause folding phenomenon (TFP product of FY-4A) will occur for a long time (at least 3 hours) above or upstream of the mixed severe convection location, which generally will also produce thunderstorm gale weather phenomena. The minute-by-minute lightning monitoring product LMIE of FY-4A and the analyzed LMIEX have no obvious indicative features for the monitoring and forecast of severe convection over the plateau.

[0134] S4. Severe convective weather forecast verification: Conduct forecast verification on the obtained forecast methods and forecast indicators during severe convective weather processes to verify the forecast lead time, accuracy rate, and data application scope of FY-4A product data in severe convective weather forecasting.

[0135] According to the forecast and early warning criteria that all conditions for severe convective potential in Table 2 need to be met simultaneously or all conditions for FY-4 products need to be met simultaneously, conduct early warning verification for the remaining severe convective processes in 2020 (see Table 3). The verification results of each forecast and early warning condition and indicator are shown in Table 4. For example, during the hail process on June 13, the maximum hail diameter near Zhidoi Station was about 7 mm at around 14:55, the maximum local hail diameter in Banma County was 4 mm at around 16:51, and the maximum hail diameter in Zadoi County was 9 mm at 18:09. The sounding structure and severe convective potential analysis chart for that day are as attached Figure 5 as shown (in Figure e, the double solid line is the 500 hPa wind shear line, the dotted coil line represents the position of the mid-upper dry layer, the small circle represents the position of the unstable energy of about 230 J / kg, and the large circle represents the position of the unstable energy greater than 1000 J / kg. The numbers are the GPS-detected atmospheric precipitable water, unit: mm). It can be seen that for Yushu Prefecture where the actual hail diameter ≥ 5 mm occurred, its CAPE was less than 600 J / kg, but there was no report of serious disasters during the severe convective weather on that day.

[0136] Table 3 Information Table for Forecast Back-substitution Verification of Severe Convective Weather Processes in 2020

[0137]

[0138] Table 4 Verification Results of Back-substitution Early Warning Application for Severe Convective Processes in 2020

[0139]

[0140]

[0141] Through the monitoring analysis and forecast and early warning verification of the severe convective process potential in 2020 and the FY-4A products, it can be obtained that currently, when using the FY-4 analysis products for severe convective early warning, the main products applicable to Qinghai are cloud type CLT, initial convective cloud CIX, cloud top temperature CTT, cloud top height CTH, instantaneous precipitation QPE, tropopause folding TFP, etc. Among them, the algorithm of the QPE product needs to be improved, and CTT and CTH are not applicable to the judgment of severe convection in areas with too high altitudes. The number of times TFP appears over severe convection in Qinghai is small, and its indication effect on severe convection still needs to be analyzed and studied using more cases. In addition, the indication effects of the AVP and LMIE products on severe convection are small. When using conventional meteorological sounding and model initial field data for severe convective potential forecasting, the conditions applicable to Qinghai are mainly sounding structure, thickness of dry and wet layers, convective available potential energy, dynamic lifting and temperature difference conditions in the middle and lower layers, etc. During the forecasting process, it is necessary to pay attention to whether the scale of the data is applicable to the scale of severe convection. In addition, due to the complex underlying surface of the plateau and sparse stations, the indication effects of vertical wind shear and 0°C layer height on severe convection are also not significant. Therefore, finally, Table 2 can be sorted into the severe convective forecast and early warning conditions shown in Table 5.

[0142] Table 5 Severe Convective Weather Forecast and Early Warning Conditions on the Plateau

[0143]

[0144]

[0145] Apply the above forecasting method to the typical severe convective processes during the flood season in 2021. For example, from June 15th to 16th, 2021, there were short-term heavy precipitation accompanied by local thunderstorm gales and hail in many places in southern Qinghai, and on August 23rd, 2021, there were hail accompanied by local thunderstorm gales in many places in northeastern Qinghai. The test results of the forecast and early warning method application are shown in Table 6.

[0146] Table 6 Test Results of Forecast and Early Warning Application for Typical Severe Convective Processes in 2021

[0147]

[0148] The present invention proposes severe convective potential forecasting and early warning indicators, as shown in Table 5; the establishment method of the severe convective potential early warning method proposed by the present invention, that is, the severe convective potential forecasting conditions in Table 5 are simultaneously satisfied, or the FY-4 analysis product early warning conditions are simultaneously satisfied, mainly adopts a method of complementary forecasting and early warning with two conditions. That is, the former can make up for the shortcoming of the relatively short early warning time of FY-4, but FY-4 can make up for the disadvantages of the low data update frequency and too coarse spatial resolution of the former. Applying the present invention to test typical severe convective processes in 2020 and 2021, the overall test results (Tables 4 and 6) show that the accuracy rate of separately conducting severe convective potential forecasting is 82%, the accuracy rate of separately conducting FY-4 severe convective early warning is 64%, and the accuracy rate of complementary early warning (almost 100% only from the test results of these typical processes) is quite high.

[0149] The above are only embodiments of the present invention, and common knowledge such as specific structures and / or characteristics known in the art are not described in detail herein. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicability of the patent. The protection scope required by this application shall be subject to the content of its claims, and the specific implementation manners and the like recorded in the specification can be used to interpret the content of the claims.

Claims

1. A method for establishing a severe convective potential warning method, characterized in that, It includes the following steps: Source data quality control: Comparing with the actual severe convective weather, detecting and removing data outliers from conventional meteorological sounding data and model initial field data, and preprocessing the FY-4 satellite source data using existing satellite calibration and positioning technologies; Parsing source data and output data: Parsing the FY-4 product source data obtained by the latest algorithm, reading the source data using the VC sharp programming language and outputting micaps format data. The output data includes FY-4 data and image products of severe convective processes; Severe convective warning methods and index extraction: Based on the actual severe convective weather, conventional meteorological sounding data, and model initial field data; Analyzing the key atmospheric parameters for severe convective potential warning, extracting severe convective warning indices, analyzing the weather characteristics of FY-4 product data before the occurrence of severe convective weather, and extracting potential warning indices; Deriving the severe convective potential warning method based on severe convective warning and potential warning indices; The FY-4 product source data includes tropopause fold (TFP), atmospheric temperature profile (AVP), convective initiation (CIX), 1-minute lightning event (LMIE), instantaneous quantitative precipitation estimation (QPE), cloud type (CLT), cloud top temperature (CTT), and cloud top height (CTH); The conventional meteorological sounding source data includes processed micaps format T_logP sounding data, wind field, temperature field, and dew point field data from the ground to high altitudes; the severe convective actual data includes automatic weather station monitoring data and manually recorded actual data; The model initial field data uses convective available potential energy (CAPE) data; Severe convective warning verification: Verifying the warning timeliness, accuracy, and data application scope of the FY-4A product data in severe convective weather warning by conducting warning verification on the obtained warning methods and warning indices during severe convective weather processes; Using the VC sharp programming language to parse and output data, and extracting warning indices and warning methods, specifically including: Stacking the LMIE within 15 minutes to process cumulative lightning monitoring data (LMIEX), and generating micaps class 3 data format from four FY-4 product data of TFP, CIX, LMIEX, and CLT; Generating micaps class 4 data format from four FY-4 product data of CTH, CTT, QPE, and TFP (including TROPPV, TROPH, Tropo_P_depth, and Tropo_Z_depth); Generating micaps class 5 format from AVP product data; Generating shp format data from TFP, CIX, and LMIEX; Generating tif format data from CTT and QPE; Importing the above-mentioned output data of FY-4 into the micaps system to analyze the performance characteristics of severe convective data and severe convective warning indices; Vertical wind shear: According to the wind field data from the ground to high altitudes, respectively calculating the vertical wind shear vector values VS0-3 and VS0-6 from the ground to 3 km height and from the ground to 6 km height; Sounding correction: Compare the location and time of the actual strong convection and the inversion situation, select the sounding station that is in the same cloud cluster as the strong convection location, with a spatial distance of no more than 150km and a sounding time before the strong convection time, and transfer the T_logP sounding data into the micaps system to display the T_logP image of the selected sounding station. If there is an inversion within 1 hour before the occurrence of strong convection, use the micaps system function to correct the T_logP image at the top of the inversion layer. If there is no inversion, use the actual ground temperature and dew point data closest to the location and time of the strong convection before the occurrence of strong convection to correct the T_logP image; Sounding structure: Output the sounding structure identification according to the corrected T_logP image; CAPE value: Calculate the estimated CAPE value based on the corrected T_logP data, compare the CAPE value of the model zero field, and take the maximum value of the two; Temperature difference between middle and lower layers T400-600: Based on the corrected T_logP data, the temperature difference between 600hPa and 400hPa where strong convection occurs is output; Dry layer thickness △Hdry and wet layer thickness △Hwet: Output the dry layer thickness and wet layer thickness of the place where severe convection occurs according to the corrected T_logP data; 0℃ layer height H0: According to the corrected T_logP data, the 0℃ layer height of the day when the severe convection occurred is output; Dynamic lifting conditions: The wind field and temperature field data from the ground to the high altitude are transferred into the micaps system to determine the wind field and temperature field in the middle and low layers where the severe convection occurs, and output the signs of whether there is wind field shear, convergence and fluctuation, and whether there is a cold temperature trough. The above conventional detection and model zero field and their output data are transferred into the micaps system to analyze the severe convection potential warning indicators; The severe convection potential warning conditions and indicators are set as: sounding structure, CAPE & vertical wind shear, dynamic lift and temperature field, middle and low layer temperature difference, dry and wet layer thickness and 0℃ layer height are met at the same time, or the atmospheric temperature profile AVP, convection initiation CIX, cloud type CLT, cloud top temperature CTT, cloud top height CTH, instantaneous quantitative precipitation estimate QPE and tropopause folding TFP are met at the same time; severe convection warning indicators are refined and severe convection warning methods are condensed.

2. The method for establishing a severe convective potential warning method according to claim 1, characterized in that: The FY-4 product data comes from the National Satellite Meteorological Center and its website, and the actual weather conditions, conventional meteorological detection data and model zero-field data come from the local meteorological bureau.

3. The method for establishing a severe convective potential warning method according to claim 1, characterized in that: The update frequency of FY-4 tropopause folded TFP data is 60 minutes, the update frequency of atmospheric temperature profile AVP and convection incipient CIX data is 15 minutes, the update frequency of 1-minute lightning event LMIE data is 1 minute, the update frequency of instantaneous quantitative precipitation estimation QPE data, cumulative lightning events LMIEX, cloud type CLT, cloud top temperature CTT, and cloud top height CTH data are all 15 minutes; the update frequency of conventional detection and model zero-field meteorological data is 12 hours.

4. The method for establishing a severe convective potential warning method according to claim 3, characterized in that: The atmospheric temperature profile AVP is specifically obtained by inverting the regional atmospheric temperature profile product from the GIIRS channel matching data set of the FY-4 satellite.

5. The method for establishing a severe convective potential warning method according to claim 1, characterized in that: Analyze the key atmospheric parameters for severe convective potential warning, including sounding structure, instability energy CAPE, wind shear, meteorological element field and circulation situation; Analyze the data performance after the analysis of FY-4 products before the occurrence of severe convective weather, where severe convective weather includes hail, short-term heavy precipitation and thunderstorm gales.

Images