Heavy rainfall nowcasting method

Abstract

The invention discloses a heavy rainfall nowcasting method, and relates to the technical field of meteorology, and the method comprises the steps: extracting original data; preprocessing the original data to obtain a multi-source data file; displaying a multi-source data file by using a Micaps platform, analyzing a wind field to identify a water vapor source and a convergence line, drawing a spatial configuration picture set of a heavy rainfall falling area, the wind field and the cloud-to-ground lightning, and constructing a heavy rainfall data set of wind field features and cloud-to-ground lightning features; based on the multi-source data file, identifying water vapor source features, wind speed fluctuation features and ground trigger mechanism features 24 hours before heavy rainfall occurs to the end; the heavy rainfall falling area is determined by utilizing the evolution characteristic that the ground-to-ground lightning density is dispersed to be dense near the ground convergence line and the area where the wind speed is changed from large to small or the wind direction is converted from the warm and wet airflow direction to the cold air direction on the water vapor source channel near the ground convergence line. According to the method, a brand new technical means is provided for heavy rainfall nowcasting by using the multi-source data with high temporal-spatial resolution, and the precision and real-time performance of forecasting are improved.

Description

Technical Field

[0001] The present invention relates to the field of meteorological technology, and in particular to a heavy rainfall nowcasting method. Background Art

[0002] Multi-source data such as regional automatic stations and lightning positioning have high temporal and spatial resolutions. The monitored features such as the ground wind field convergence zone and the ground-to-ground lightning intensive zone are of great significance to the intensity, fall area and evolution of heavy precipitation. Existing studies have shown that lightning data can be obtained at any time and has a wide coverage, which can improve satellite precipitation estimates and better understand and predict convective storms that occur in flash floods. Severe weather tends to occur during the development and maturity of storms and in lightning activity areas to the south and west of convective systems. In the mature stage, negative ground-to-ground lightning gathers in large numbers and reaches a peak, then decreases. In the decay stage, negative ground-to-ground lightning decreases rapidly but still gathers in the convective area. The area with high frequency of negative ground-to-ground lightning appears in the area of ​​enhanced convective precipitation. The linear correlation between lightning activity and precipitation in the convective activity area is significant. Ground-to-ground lightning is prone to occur in more organized and stronger convective systems. The mesoscale convergence lines in the regional automatic station wind field appear before heavy rainfall, and the local convergence caused by the convergence lines is a mesoscale system that initiates convective activity. In addition, the mesocyclones, medium and low pressure, and small and medium-scale cyclonic convergence wind fields and shear lines observed by the automatic meteorological station induce the development of convective systems. When the convective systems develop to the mature stage, the density of ground flashes reaches the maximum value.

[0003] At present, the forecast of heavy rainfall in forecasting business mainly relies on numerical forecast products, but due to the local, sudden and high-intensity characteristics of heavy rainfall, the objective forecasting method of numerical forecasting cannot capture the sudden local heavy rainfall area; in addition, there are many research results on the formation mechanism of heavy rainfall in forecasting business, but there are few reports on the nowcasting method that uses the combination of high-temporal and spatial ground wind field data and lightning location data to capture the early signal characteristics before the occurrence of heavy rainfall, and then determine the heavy rainfall area. Summary of the invention

[0004] The present invention provides a heavy rainfall nowcasting method, which can solve the above-mentioned problem.

[0005] In order to solve the above problems, the technical solution adopted by the present invention is as follows:

[0006] The invention provides a heavy rainfall nowcasting method, comprising: extracting raw data, including hourly precipitation data, hourly 2-minute average wind data and hourly lightning location data; preprocessing the raw data to obtain a multi-source data file, including a multi-element mapping data file, a ground full-element mapping data file and a lightning location data file; using a Micaps platform to display the multi-source data file, analyzing the wind field for water vapor sources and convergence lines, drawing a spatial configuration picture set of heavy rainfall areas, wind fields and ground lightning, and constructing a heavy rainfall data set of wind field characteristics and ground lightning characteristics; based on the multi-source data file, identifying water vapor source characteristics, wind speed pulsation characteristics and ground triggering mechanism characteristics from 24 hours before the occurrence of heavy rainfall to the end, and determining the heavy rainfall area by using the evolution characteristics of ground lightning density from sparse to dense near the ground convergence line, and the area where the wind speed changes from large to small or the wind direction changes from the direction of warm and humid air flow to the direction of cold air on the water vapor source channel near the ground convergence line.

[0007] As a further description of the above technical solution, program code is written in C# language to extract raw data from the "Meteorological Big Data Cloud Platform Tianqing" and store it in txt text format; the fields of hourly precipitation data include province, city (county, banner, township), station name, station number, longitude, latitude, altitude, precipitation and precipitation quality control code; the fields of hourly 2-minute average wind data include province, city (county, banner, township), station name, station number, longitude, latitude, altitude, 2-minute average wind direction, 2-minute average wind speed, 2-minute average wind direction quality control code and 2-minute average wind speed quality control code; the fields of hourly lightning location data include record identification, data source, data identification, storage time, receipt time, update time, data time, latitude, longitude, positioning error, current (return stroke peak) intensity, maximum return stroke steepness, positioning method, lightning geographical location information province, lightning geographical location information city, lightning geographical location information county and correction report mark.

[0008] As a further description of the above technical solution, the method for processing hourly precipitation data includes: using C# language to write program code to extract data that meets the heavy precipitation standard (1 hour precipitation is greater than or equal to 20 mm) and has a precipitation quality control code of 0 from the hourly precipitation data, and converting the hourly precipitation data after quality control into Micaps 34th type data format (referred to as diamond 34 type data), that is, obtaining a multi-factor mapping data file;

[0009] The start time is recorded when heavy precipitation occurs at more than one station in all the hourly precipitation data after quality control at the same time, and the end time is recorded when heavy precipitation occurs for more than two consecutive moments after the heavy precipitation occurs. The heavy precipitation case data are extracted from the start time to the end time, and the fields include time, station name, station number, longitude, latitude, and hourly precipitation. The heavy precipitation data set table is constructed using the start time and end time of the heavy precipitation case information, and the table fields include case number, start time, end time, extreme rainfall intensity, station, ground lightning characteristics, and ground wind field information.

[0010] As a further description of the above technical solution, a method for processing hourly 2-minute average wind data includes: using C# language to write program code to extract the 2-minute average wind direction quality control code and the 2-minute average wind speed quality control code of 0 from the 2-minute average wind data, and convert them into Micaps Class 1 data format (referred to as Diamond Class 1 data), that is, to obtain a ground full-element mapping data file; the time period to which the ground full-element mapping data file belongs is from 24 hours before the occurrence of heavy rainfall to the end of heavy rainfall.

[0011] As a further description of the above technical solution, a method for processing hourly lightning location data is: using C# language to write program code to convert the hourly lightning location data into Micaps Class 41 data format (referred to as diamond 41 data), that is, to obtain a lightning location data file; the time period to which the lightning location data file belongs is from 24 hours before the occurrence of heavy rainfall to the end of heavy rainfall.

[0012] As a further description of the above technical solution, the identification process of water vapor source characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case, calling the diamond 34 data, diamond 1 data and diamond 41 data from 24 hours before the heavy rainfall to the same time of the end on the Micaps platform; analyzing the water vapor source through the wind field, filling in the water vapor source wind direction and wind speed evolution characteristics from 24 hours before the heavy rainfall to the end into the corresponding case in the heavy rainfall data set table one by one.

[0013] As a further description of the above technical solution, the identification process of wind speed pulsation characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case, calling the diamond 34 type data, diamond 1 type data, and diamond 41 type data from 24 hours before the heavy rainfall to the same time of the end on the Micaps platform to analyze the wind speed pulsation characteristics; filling the wind speed pulsation analysis results from 24 hours before the heavy rainfall to the end into the corresponding cases in the heavy rainfall data set table one by one.

[0014] As a further description of the above technical solution, there are two types of wind speed pulsation characteristics. One is that the wind speed in the water vapor channel in the same area has increased compared with the previous period, and the other is that the wind speed in the water vapor channel has increased from the sea surface to the heavy precipitation area. Heavy precipitation is prone to occur in areas where the wind speed in the water vapor source channel near the ground convergence line changes from large to small or the wind direction changes from the direction of warm and humid air flow to the direction of cold air.

[0015] As a further description of the above technical solution, the identification process of the ground triggering mechanism characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case set, calling the diamond34 type data, diamond1 type data, and diamond41 type data from 24 hours before the occurrence of heavy rainfall to the same time of the end on the Micaps platform; analyzing the position and movement characteristics of the convergence line from 24 hours before the occurrence of heavy rainfall to the end; filling the analysis results of the appearance time and position evolution characteristics of the convergence line from 24 hours before the occurrence of heavy rainfall to the end into the corresponding cases in the heavy rainfall data set table one by one; analyzing the evolution characteristics and density of ground-to-ground lightning activity near the convergence line; filling the dense evolution characteristics of ground-to-ground lightning from 24 hours before the occurrence of heavy rainfall to the end, and the configuration relationship between the dense ground-to-ground lightning area and the heavy rainfall into the corresponding cases in the heavy rainfall data set table one by one.

[0016] As a further description of the above technical solution, according to the occurrence time of heavy rainfall in individual heavy rainfall cases, the Micaps platform calls diamond 34 data, diamond 1 data, and diamond 41 data from 24 hours before the occurrence of heavy rainfall to the same time when it ends to draw a set of spatial configuration pictures of heavy rainfall areas, wind fields, and ground flashes.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] 1) Combining multi-source data files to identify and analyze the sources of water vapor and convergence lines in the wind field before heavy rainfall and the evolution characteristics of ground lightning, it can effectively capture the key indicator characteristics before the occurrence of local heavy rainfall, such as the convergence lines of the ground wind field, the dense area of ​​ground lightning and its evolution characteristics over time, thereby improving the timeliness and accuracy of the heavy rainfall area and intensity forecast.

[0019] 2) The use of high temporal and spatial resolution ground wind field and lightning data, combined with the accurate identification of wind speed pulsation characteristics, water vapor source characteristics and ground triggering mechanism characteristics, can effectively make up for the deficiency of numerical forecasting in timely capturing local heavy rainfall areas.

[0020] 3) The hourly precipitation data, 2-minute average wind data and lightning location data were integrated and preprocessed to form a multi-source data file. Through multi-element mapping and ground full-element mapping, a set of spatial configuration pictures of heavy rainfall areas, wind fields and ground lightning were drawn. This can not only intuitively reflect the evolution characteristics before heavy rainfall, but also provide a more reliable basis for judgment for nowcasting.

[0021] 4) A heavy precipitation dataset was constructed based on wind field characteristics and ground-to-ground lightning characteristics, which quantified the linear correlation between wind field, ground-to-ground lightning and heavy precipitation. This can capture the evolution process of local severe convective systems and provide a data basis for the subsequent optimization of heavy precipitation forecast models. It also provides a new data analysis method for the study of the mechanism of heavy precipitation.

[0022] 5) By utilizing the evolution characteristics of ground wind convergence lines and ground lightning density from sparse to dense, combined with the ground trigger mechanism, the heavy rainfall area can be quickly identified, which can significantly shorten the warning time before the occurrence of heavy rainfall, provide important technical support for disaster prevention and mitigation, and have high application value.

[0023] 6) The regional automatic station data and lightning location data used have a wide coverage, high resolution, and strong real-time performance. They can be widely used in the monitoring and nowcasting of local heavy rainfall, and are especially suitable for the rapid analysis and real-time forecasting of small and medium-scale convective weather.

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 is a flow chart of the heavy rainfall nowcasting method described in the embodiment;

[0027] Figure 2 is a flow chart of a water vapor identification method in an embodiment;

[0028] Figure 3 It is the sea level wind field (wind feather) of the heavy precipitation case 34 in the Nenjiang River Basin in the embodiment (the heavy precipitation occurred between 15:00 and 16:00 on August 23, 2021), ground flash (positive and negative signs represent positive flash and negative flash, respectively), heavy precipitation falling area (the value represents the rain intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line;

[0029] Figure 4It is the sea level wind field (wind feather) of the heavy precipitation case 42 in the Nenjiang River Basin in the embodiment (the heavy precipitation occurred between 14:00 and 17:00 on June 30, 2022), ground flash (positive and negative signs represent positive flash and negative flash, respectively), heavy precipitation falling area (the value represents the rainfall intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line;

[0030] Figure 5 It is the sea level wind field (wind feather) of the heavy precipitation case 6 in the Nenjiang River Basin in the embodiment (the heavy precipitation occurred on July 28, 2020, 09:00-11:00), ground flash (positive and negative signs represent positive flash and negative flash, respectively), heavy precipitation area (the value represents the rainfall intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line;

[0031] Figure 6 is a flow chart of wind speed pulsation feature recognition in an embodiment;

[0032] Figure 7 It is the sea level wind field (wind feather) of the heavy precipitation case 81 in the Nenjiang River Basin in the embodiment (the heavy precipitation occurred from 21:00 on August 2 to 10:00 on August 3, 2023), ground flash (positive and negative signs represent positive flash and negative flash respectively), heavy precipitation falling area (the value represents the rainfall intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line;

[0033] Figure 8 is a flow chart of ground trigger mechanism feature identification in an embodiment;

[0034] Fig. 9 It is the sea level wind field (wind feather) of the heavy precipitation case 82 in the Nenjiang River Basin in the embodiment (the heavy precipitation occurred at 16:00 on August 3, 2023), ground flash (positive and negative signs represent positive flash and negative flash respectively), heavy precipitation falling area (the value represents the rain intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line;

[0035] Fig.10 It is the sea level wind field (wind feather) of heavy rainfall case 83 in the Nenjiang River Basin in the embodiment (the heavy rainfall occurred between 20:00 on August 3 and 00:00 on August 4, 2023), ground flash (the positive and negative signs represent positive flash and negative flash, respectively), heavy rainfall area (the numerical value represents rainfall intensity, unit: mm), the black arrow curve represents the source of ground water vapor, and the cross-straight line represents the ground convergence line. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0037] The embodiment of the present invention provides a heavy rainfall nowcasting method, which is specifically as follows:

[0038] 1. Data extraction

[0039] Using C# language, the hourly precipitation data and hourly 2-minute average wind data (i.e. 2-minute average wind data recorded once an hour) from 20:00 to 20:00 the next day (Beijing time, the same below) in the Nenjiang River Basin (120-128°E east longitude, 45-52°N north latitude), including national stations and automatic weather stations in eastern Inner Mongolia, Heilongjiang, and Jilin from May to September 2020-2023 were extracted through the "Meteorological Big Data Cloud Platform·Tianqing". A total of 1,099 regional national stations and automatic weather stations were selected.

[0040] The hourly precipitation data fields include province, city (county, banner, township), station name, station number, longitude, latitude, altitude, precipitation, and precipitation quality control code.

[0041] The hourly 2-minute average wind data fields include province, city (county, banner, township), station name, station number, longitude, latitude, altitude, 2-minute average wind direction, 2-minute average wind speed, 2-minute average wind direction quality control code, and 2-minute average wind speed quality control code.

[0042] Extract hourly lightning location data (i.e. lightning location data collected every hour). The fields include record ID, data source, data ID, storage time, receipt time, update time, data time, latitude, longitude, positioning error, current (return stroke peak) intensity, maximum return stroke steepness, positioning method, lightning geographic location information province, lightning geographic location information city, lightning geographic location information county, and correction report mark.

[0043] The above data are stored in txt text format.

[0044] 2. Data processing

[0045] 1) Precipitation data processing

[0046] Heavy precipitation is defined as precipitation greater than or equal to 20 mm in 1 hour.

[0047] Using C# language, program code is written to process the raw hourly precipitation data, and data that meets the heavy precipitation standard and has a precipitation quality control code of 0 is extracted from the hourly precipitation data. The fields include province, city (county, banner, township), station name, station number, longitude, latitude, altitude, and precipitation. This data is called hourly precipitation data after quality control and is stored in txt file format.

[0048] Using C# language, program code was written to extract 89 cases of heavy precipitation from the hourly precipitation data after quality control; using C# language, the heavy precipitation case data was converted into Micaps Class 34 (multi-factor mapping) data format, and the file was named after the date and time when the heavy precipitation occurred, such as the data at 12:00 on July 9, 2023, the file was named PRE-2023070912.txt, and the file content was as follows:

[0049] Diamond 34 Precipitation at 12:00 on July 9, 2023 2023 07 09 12 5000 500 3

[0053] Sequence 1 Sequence 2 Sequence 3

[0054] Numbers Characters Numbers

[0055] E6218 123.91 45.8 21.5 21.5 21.5

[0056] C0127 120.08 49.96 20.5 20.5 20.5 ......

[0058] 2) Hourly 2-minute average wind data processing

[0059] Using C# language, program code is written to process the hourly 2-minute average wind speed (referred to as wind speed, the same below) and 2-minute average wind direction (referred to as wind direction, the same below) data, and extract the 2-minute average wind data with the 2-minute average wind direction quality control code and the 2-minute average wind speed quality control code both being 0. The fields include province, city, county (banner, township), station name, station number, longitude, latitude, altitude, wind direction, and wind speed. This data is called the 2-minute average wind data after quality control and is stored in txt file format.

[0060] Using C# language, write program code to convert the 2-minute average wind data after quality control into Micaps Class 1 (ground full-element mapping) data format. The generated data format file period is from 24 hours before the heavy rainfall to the end of the heavy rainfall. The file is named after the date and time when the heavy rainfall occurs. For example, the data at 12:00 on July 9, 2023 is named WIND-2023070912.txt. The file content is as follows:

[0061] diamond 1 Surface wind field at 12:00 on July 9, 2024 2024 07 09 12 5000 53192 115 44.02 0 1 0 170 2.2 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 50618 118.26 48.19 0 1 0 197 2.9 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 54012 117.63 44.57 0 1 0 98 2.6 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 50834 121.21 46.6 0 1 0 232 2.4 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999 ......

[0068] 3) Hourly lightning location data processing

[0069] Using C# language, program code is written to process the hourly lightning location data, and the lightning location data is converted into Micaps Class 41 (lightning location) data format. The generated data format file period is from 24 hours before the heavy rainfall to the end of the heavy rainfall. The file is named after the date and time when the heavy rainfall occurs. For example, the data at 12:00 on July 9, 2023 is named LIGHT-2023070912.txt, and the file content is as follows:

[0070] Diamond 41 2023-07-09 12 Lightning Monitoring Data 2023 07 09 12 10000

[0072] 1 20237901211855820 9999 9999 120.8701 50.4404 9999 9999 9999 9999 -63.9664 99990 2 -10.2346

[0073] 2 202379012127906623 9999 9999 117.2819 49.8935 9999 9999 9999 9999 -32.231 99990 2 -4.0446

[0074] 3 202379012131283776 9999 9999 123.0627 48.4073 9999 9999 9999 9999 -85.915 99990 2 -10.2776

[0075] 4 2023790121567755527 9999 9999 116.7449 47.0968 9999 9999 9999 9999 -82.7518 9999 0 2 -16.813

[0076] 5 2023790122238197988 9999 9999 117.1861 49.8991 9999 9999 9999 9999 -100.7594 9999 0 2 -13.296 ......

[0078] 3. Heavy precipitation analysis

[0079] Using the Micaps platform, multi-factor mapping data files (for example, PRE-2023070912.txt file), ground full-factor mapping data files (for example, WIND-2023070912.txt file), and lightning location data files (for example, LIGHT-2023070912.txt file) were retrieved from 24 hours before the heavy rainfall to the end of the heavy rainfall. The wind field convergence lines, cyclones, and water vapor sources were analyzed through the Micaps platform interactive tools. At the same time, a set of spatial configuration pictures of heavy rainfall areas, wind fields, and ground flashes were drawn and saved as pictures named by date and time. A total of 1,290 picture files were generated. Combined with meteorological principles, 89 heavy rainfall cases in the Nenjiang River Basin from May to September 2020 to 2023 were analyzed from the wind direction and wind speed to analyze the source of heavy rainfall water vapor, wind speed pulsation, ground convergence, and the correspondence between the evolution of ground lightning density and the heavy rainfall area. A heavy rainfall dataset with wind field characteristics and ground lightning characteristics was constructed. The fields include case number, start time, end time, rainfall intensity extreme value, station, ground lightning characteristics, and wind field characteristics. Specifically, it includes:

[0080] (1) Identification of water vapor source characteristics

[0081] refer to Figure 2 As shown, the Micaps platform calls diamond 34 data, diamond 1 data, and diamond 41 data at the same time; the source of water vapor is analyzed through the wind field.

[0082] Among them, the southeast wind from the Sea of ​​Japan is more concentrated, which is defined as the near-ground water vapor coming from the Sea of ​​Japan (referred to as the water vapor coming from the Sea of ​​Japan); the south wind from the Bohai Sea is more concentrated, which is defined as the near-ground water vapor coming from the Bohai Sea (referred to as the water vapor coming from the Bohai Sea); if the water vapor comes from the Bohai Sea (Sea of ​​Japan) and the Sea of ​​Japan (Bohai Sea) at the same time or successively, it is defined as the water vapor coming from the Sea of ​​Japan and the Bohai Sea.

[0083] Wind direction from Bohai Sea (Sea of ​​Japan) is mostly winds other than southerly and southeasterly or easterly, which is defined as having no obvious source of water vapor.

[0084] Fill in the evolution characteristics of wind direction and wind speed of water vapor sources from 24 hours before the heavy rainfall to the end into the corresponding cases in the data set table one by one.

[0085] According to the analysis of individual cases, among the 89 cases, the near-ground water vapor came from the Bohai Sea, the Sea of ​​Japan, and both the Bohai Sea and the Sea of ​​Japan in 36, 33, and 17 cases respectively before the heavy rainfall occurred, and there was no obvious source of water vapor in the other 3 cases.

[0086] The conclusion of water vapor identification is: before the heavy rainfall identified by the wind field occurs, there must be water vapor sources from the Bohai Sea, the Sea of ​​Japan, or both the Bohai Sea and the Sea of ​​Japan as a necessary condition for the occurrence of heavy rainfall.

[0087] 1) Water vapor comes from Bohai Sea Case: From 15:00 to 16:00 on August 23, 2021 (Case 34), heavy rainfall occurred in the lower reaches of the Nuomin River in the Nenjiang River Basin, which lasted for 2 hours. The maximum rainfall intensity occurred at 15:00, which was 24.9 mm. The sea level wind field and ground lightning showed that 24 hours before the heavy rainfall occurred, at 15:00 on August 22, 2021, the Nenjiang River Basin was northwest wind ( Figure 3 a), wind speed is within 4m / s, from Liaoning to the lower reaches of Nenjiang River, the wind speed is relatively high in Liaoning, with the maximum reaching 8m / s, and only a small amount of ground lightning activity occurs in the lower reaches of Nenjiang River; 16h before the heavy rainfall, at 23:00 on August 22, 2021, the wind field characteristics have little change ( Figure 3 b), but ground-to-ground lightning activity began to increase, especially in the lower reaches of the Nenjiang River Basin; 13 hours before the heavy rainfall, at 02:00 on August 23, 2021, the wind from Liaoning to the lower reaches of the Nenjiang River was still southerly ( Figure 3 c) In Liaoning, the wind speed was relatively strong, reaching a maximum of 8 m / s, and the dense ground lightning area moved eastward to the Songnen Plain; 4 hours before the heavy rainfall, at 11:00 on August 23, 2021, the wind speed increased significantly ( Figure 3 d), the maximum wind speed from Liaoning to central Jilin reached 8m / s, and ground lightning was frequent in the Nenjiang River Basin; at 15:00 on August 23, 2021, the maximum wind speed from Liaoning to central Jilin reached 8m / s ( Figure 3 e) There were frequent ground lightning strikes in the lower reaches of Ganhe River in the Nenjiang River Basin, corresponding to the heavy rainfall area; at 16:00 on August 23, 2021, the wind speed in Liaoning decreased to within 4m / s ( Figure 3 f), but the maximum wind speed in central Jilin still reached 8m / s, the frequency of ground lightning in the heavy rainfall area decreased compared with the previous moment, and the area with intensive ground lightning still corresponded to heavy rainfall weather.

[0088] 2) Water vapor comes from the Sea of ​​Japan: From 14:00 to 17:00 on June 30, 2022, heavy precipitation occurred in the lower reaches of the Nuomin River in the Nenjiang River Basin (Case 42), lasting 4 hours, with a maximum rainfall intensity of 36.8 mm. The sea level wind field and ground lightning show that 24 hours before the heavy precipitation occurred, at 14:00 on June 29, 2022, the wind from Liaoning to the Nenjiang River Basin was northwest wind ( Figure 4 a), wind speed within 4m / s, easterly wind from the Sea of ​​Japan to Heilongjiang, wind speed within 4m / s, more ground lightning on the west bank of the Nenjiang River Basin; 18h before the heavy rainfall, at 20:00 on June 29, 2022, the wind field characteristics changed little ( Figure 4 b), but the ground-to-ground lightning activity began to decrease, and the ground-to-ground lightning activity was more intensive in the middle reaches of the west bank of the Nenjiang River Basin; 12 hours before the heavy rainfall, at 02:00 on June 30, 2022, the wind from Liaoning to the lower reaches of the Nenjiang River was still northwest or north ( Figure 4c), the wind speed is within 4m / s, but the wind speed from the Sea of ​​Japan to Heilongjiang increases significantly, with the maximum wind speed reaching 8m / s, and the dense area of ​​ground lightning moves eastward; 6h before the heavy rainfall occurs, at 08:00 on June 30, 2022, the wind speed from the Sea of ​​Japan to Heilongjiang decreases, basically within 6m / s ( Figure 4 d), the ground-to-ground lightning activity in the Nenjiang River Basin is not obvious; at 14:00 on June 30, 2022, the wind from the Sea of ​​Japan to Heilongjiang is still easterly ( Figure 4 e), wind speed within 6m / s, there are dense ground lightning in the lower reaches of Ganhe River in the Nenjiang River Basin, corresponding to the heavy rainfall area; at 17:00 on June 30, 2022, the wind from the Sea of ​​Japan to Heilongjiang is still easterly ( Figure 4 f), within the wind speed of 6m / s, the frequency of ground flashes in the heavy rainfall area decreased compared with the previous moment, but the area with dense ground flashes still corresponded to heavy rainfall weather. In summary, from 24h before the heavy rainfall to the time of the occurrence, there was an easterly airflow near the ground layer, the water vapor came from the Sea of ​​Japan, and the area near the ground convergence line where ground flashes changed from sparse to dense corresponded to the heavy rainfall area.

[0089] 3) Water vapor comes from the Bohai Sea and the Sea of ​​Japan. Case 6: From 09:00 to 11:00 on July 28, 2020, heavy rainfall lasted for 3 hours in the middle reaches of the Nenjiang River, with a maximum rainfall intensity of 75.5 mm. The sea level wind field and ground lightning showed that 24 hours before the heavy rainfall occurred, at 09:00 on July 27, 2020, the wind from Liaoning to the Nenjiang River Basin was southerly ( Figure 5 a), the wind from the Sea of ​​Japan to Heilongjiang is easterly, with wind speeds within 4m / s. Two streams of water vapor converge on the east bank of the Nenjiang River Basin, and there is basically no ground lightning activity in the Nenjiang River Basin; 15h before the heavy rainfall, at 20:00 on July 27, 2020, the wind field characteristics have little change ( Figure 5 b), no ground lightning activity; 7h and 4h before the heavy rainfall, at 02:00 on July 28, 2020 ( Figure 5 c) and 05:00 on July 28, 2020 ( Figure 5 d), the wind direction changes little, and the water vapor from the Sea of ​​Japan and Bohai Sea is still gathering in the Nenjiang River Basin in the near-ground layer. The wind speed is within 4m / s, and there is still no ground lightning activity; when heavy rainfall occurs, at 09:00 on July 28, 2020, the wind from the Sea of ​​Japan to Heilongjiang is still easterly ( Figure 5 e), the wind from Bohai Sea to Nenjiang River Basin is still southerly, and two water vapors converge in the lower reaches of Yalu River in Nenjiang River Basin. Heavy precipitation occurs in the convergence area of ​​southerly and easterly winds, and ground lightning is dense in the northwest of the heavy precipitation area; at 11:00 on July 28, 2020, the wind direction changes little ( Figure 5 f), the heavy rainfall area is located in the south of the dense ground lightning area. In summary, from 24 hours before the heavy rainfall to the time of the heavy rainfall, there were easterly and southerly airflows near the ground, and the water vapor came from the Sea of ​​Japan and the Bohai Sea. Heavy rainfall occurred in the convergence area of ​​the southerly and easterly winds. In particular, this case occurred at 09:00 on July 28, 2020 ( Figure 5e) The wind speed on the east side of the heavy rainfall area is 4m / s, but the wind speed in the heavy rainfall area is significantly reduced to 2m / s, which proves that heavy rainfall is prone to occur in areas where the wind speed of the water vapor source channel decreases.

[0090] (2) Wind speed pulsation feature recognition

[0091] The process of wind speed pulsation feature recognition is as follows: Figure 6 shown.

[0092] Wind speed pulsation is defined as the phenomenon that the wind direction in a certain area is consistent and the wind speed increases over time.

[0093] There are two characteristics in the identification of wind speed pulsation: the wind speed of the water vapor channel in the same area has increased compared with the previous period, and the wind speed of the water vapor channel has increased from the sea to the heavy rainfall area.

[0094] The Micaps platform calls diamond 34 data, diamond 1 data, and diamond 41 data at the same time.

[0095] From 24 hours before the occurrence of heavy rainfall to the end, the wind speed pulsation characteristics are analyzed through the wind field wind speed.

[0096] Fill in the wind speed pulsation analysis results from 24 hours before the heavy rainfall to the end of the rainfall into the corresponding cases in the data set table one by one.

[0097] According to the case analysis, among the 89 cases, 57 cases (71.25%) showed wind speed pulsation characteristics in the wind field of the water vapor source before the heavy precipitation occurred. There are two main times for wind speed pulsation: one is before the heavy precipitation occurs, and the other is when the heavy precipitation occurs.

[0098] The conclusion of wind speed pulsation identification is: wind speed pulsation occurs before heavy precipitation or during continuous heavy precipitation, which increases water vapor transport and is conducive to the occurrence of heavy precipitation.

[0099] Cases where the wind speed continued to increase before and after the heavy rainfall: From 21:00 on August 2, 2023 to 10:00 on August 3, heavy rainfall lasting 14 hours occurred on the west bank of the middle and lower reaches of the Nenjiang River Basin (Case 81), with a maximum rainfall intensity of 76.2 mm. The heavy rainfall first appeared at 21:00 on August 2, in the middle and upper reaches of the Taoer River on the west bank of the Nenjiang River to the Chuoer River area in the south, and lasted until 04:00 on August 3. After that, the heavy rainfall area moved north to the middle and lower reaches of the Yalu River on the west bank of the middle reaches of the Nenjiang River Basin, and lasted until 10:00 on August 3. The sea level wind field and ground lightning show that 24 hours before the heavy rainfall occurred, at 21:00 on August 1, 2023, there was a southerly wind from the Bohai Sea to the east bank of the Nenjiang River Basin, and only in Liaoning the maximum wind speed reached 6m / s ( Figure 7a), there is no ground lightning in the Nenjiang River Basin; at 05:00 on August 2, 2023, the wind from Bohai to the east coast of the Nenjiang River Basin is still southerly, and the maximum wind speed in Liaoning increases to 8m / s ( Figure 7 b), dense ground lightning began to appear south of Taoer River in the lower reaches of Nenjiang River; at 11:00 on August 2, 2023, the wind from Bohai Sea to the east coast of Nenjiang River Basin was still southerly, and the area with the maximum wind speed in Liaoning increased to 8m / s and expanded northward to the southern part of Jilin ( Figure 7 c) There is no ground lightning in the Nenjiang River Basin; at 14:00 on August 2, 2023, the wind from Bohai to the east coast of the Nenjiang River Basin is still southerly, and the wind speed in Liaoning continues to increase, with the maximum wind speed increasing to 10m / s ( Figure 7 d), dense ground lightning began to appear south of Taoer River in the lower reaches of Nenjiang River; from 15:00 to 21:00 on August 2, 2023, the wind from Bohai to the east coast of Nenjiang River is still southerly, and the maximum wind speed from Liaoning to southern Jilin is still 8m / s (figure omitted), and heavy precipitation began to appear at 21:00; at 23:00 on August 2, 2023, the southerly wind from Bohai Sea decreased in Jilin, but the maximum wind speed in Liaoning was still 8m / s. At the same time, the easterly airflow from the Sea of ​​Japan has been established, but the wind speed is within 4m / s. The dense ground lightning area overlaps with the heavy precipitation area ( Figure 7 e); At 02:00 on August 3, 2023, the maximum wind speed of the south wind from Bohai Sea in Liaoning is still 8m / s, and the wind speed of the east wind from the Sea of ​​Japan is within 4m / s. The dense ground lightning area overlaps with the heavy rainfall area ( Figure 7 f); 02:00 to 08:00 on August 3, 2023 ( Figure 7 g), Liaoning's wind speed has little change, but Heilongjiang's northeast or east wind speed begins to increase, with a maximum wind speed of 6m / s, but ground lightning has decreased significantly; August 3, 2023 10:00 ( Figure 7 h), the wind speed from Liaoning to southern Jilin increased, reaching a maximum of 10m / s, the easterly wind from the Sea of ​​Japan was 4m / s, there was no ground lightning, and the heavy rainfall was coming to an end.

[0100] In summary, from 24 hours before to the time of the heavy rainfall, water vapor came from the Bohai Sea, and the wind speed in the near-ground layer from the Bohai Sea showed a characteristic of increasing twice before the heavy rainfall occurred, and increased forwardly towards the heavy rainfall area, with the maximum wind speed increasing to 10m / s, creating favorable water vapor transport conditions for the occurrence of heavy rainfall. The heavy rainfall occurred in the area of ​​reduced wind speed in the dense ground lightning area near the convergence line.

[0101] (3) Ground trigger mechanism feature identification

[0102] The ground trigger mechanism feature identification process is as follows Figure 8 shown.

[0103] Convergence line definition: the area where the wind direction on the water vapor source channel changes from the direction of warm and humid air flow to the direction of cold air; mesocyclone definition: a closed cyclonic circulation in the wind field is called a mesocyclone; secondary cold front definition: a front that is generated behind a cold front and has similar properties to a cold front; wind speed convergence is defined as an area where the wind speed in the direction of the water vapor source near the convergence line changes from large to small;

[0104] The Micaps platform calls diamond 34 data, diamond 1 data, and diamond 41 data at the same time;

[0105] Analyze the position and movement characteristics of the convergence line (cyclone, secondary cold front) from 24 hours before the heavy rainfall to the end; fill in the analysis results of the appearance time and falling area of ​​the convergence line (cyclone, secondary cold front) from 24 hours before the heavy rainfall to the end in the corresponding cases in the data set table one by one;

[0106] Analyze the evolution characteristics of ground-to-ground lightning activity near the convergence line and whether it is dense;

[0107] Fill in the corresponding cases in the data set table one by one with the evolution characteristics of ground-to-ground lightning density from 24 hours before the occurrence of heavy rainfall to the end, and the relationship between the ground-to-ground lightning density area and heavy rainfall configuration;

[0108] According to the case analysis, the triggering mechanisms of the near-surface wind field in 89 cases mainly include ground convergence line, cyclone convergence center (case 35), secondary cold front, wind speed convergence, wind speed and wind direction convergence (case 9);

[0109] In 89 cases, there were ground convergence lines before or during heavy rainfall. In 68 cases, the lines appeared 24 hours before the heavy rainfall (accounting for 76.4%). In 6 cases, the lines appeared 21-23 hours before the heavy rainfall (case 2, case 3, case 5, case 11, case 58, case 88). In 10 cases, the lines appeared 12-19 hours before the heavy rainfall (case 14, case 33, case 37, case 47, case 49, case 51, case 57, case 69, case 75, case 76). In 3 cases, the lines appeared 7-9 hours before the heavy rainfall (case 6, case 36, case 87). In 2 cases, the lines appeared during the heavy rainfall (case 17, case 74).

[0110] The main shapes of the convergence lines are north-south, southeast-northwest, and "human" shapes (case 28, case 81).

[0111] The characteristics of the convergence line are mainly stable and less moving, swinging from north to south, moving in the northeast direction, and moving in the southeast direction. There are also a few cases where the wind speed converges in the later stage of the convergence line in the early stage (case 18, case 26), and the convergence line appears when heavy rainfall occurs (case 74);

[0112] The convergence line and the secondary cold front or cyclone convergence center have the characteristics of transition, alternation or co-appearance;

[0113] The conclusion of the triggering mechanism identification is that the vicinity of the ground convergence line (cyclone, secondary cold front) is the area where heavy precipitation occurs, and heavy precipitation will occur near the area where ground flashes change from scattered to dense; the ground convergence line or cyclone advances from south to north, indicating that the warm and humid air flow is strong, and heavy precipitation of large magnitude is prone to occur; the speed of the convergence line or cyclone advancing northward slows down or stagnates, indicating that the strength of the cold and warm air is equal, and at this time, heavy precipitation of large magnitude will occur in the area with dense ground flashes near the convergence line or cyclone; the convergence line moves to the southeast or south, indicating the invasion of cold air, and heavy precipitation will occur in the area with dense ground flashes near the convergence line.

[0114] Cases of ground convergence lines: From August 2 to 4, 2023, continuous heavy rainfall occurred in the middle and lower reaches of the Nenjiang River Basin. During this period, from 21:00 on August 2 to 10:00 on August 3, 2023 (case 81), 16:00 on August 3, 2023 (case 82), and from 20:00 on August 3 to 00:00 on August 4, 2023 (case 83), due to the joint influence of the alternating and continuous appearance of convergence lines and "herringbone" convergence lines, the intervals between heavy rainfall cases were only 6 hours and 4 hours. The specific analysis is as follows:

[0115] 6 hours after the heavy precipitation case 81, at 16:00 on August 3, 2023 (case 82), heavy precipitation weather lasting 3 hours occurred in the middle and lower reaches of the Nenjiang River Basin, with a maximum rainfall intensity of 49.1 mm. Before this heavy precipitation case occurred, sea vapor from the Bohai Sea and the Sea of ​​Japan converged in the Nenjiang River Basin. At 14:00 on August 3, 2023, a south-north convergence line was formed on the west bank of the Nenjiang River Basin under the influence of the easterly airflow ( Fig. 9 a), affected by the southerly wind from Bohai, a southwest-northeast convergence line was formed in the northwest of Jilin, and heavy precipitation occurred on the northern convergence line, but there was no obvious ground lightning activity in the heavy precipitation area; at 16:00 on August 3, 2023, the northerly wind in the Nenjiang River Basin pressed south, and under its influence, the two convergence lines moved to the south ( Fig. 9 b) Affected by the southerly air current from the Bohai Sea, a southwest-northeast convergence line was formed in the northwest of Jilin. Heavy rainfall occurred on the northern convergence line and sporadic ground lightning activities began to appear.

[0116] 4 hours after the heavy precipitation case 81, from 20:00 on August 3, 2023 to 00:00 on August 4, 2023 (case 83), heavy precipitation weather lasting for 5 hours occurred in the middle and lower reaches of the Nenjiang River Basin, with a maximum rainfall intensity of 61.2 mm. At 20:00 on August 3, 2023, the wind direction from the Nenjiang River Basin to the vicinity of the Heilongjiang River turned to north or northeast wind ( Fig.10a), a convergence line is formed. The wind from Bohai to southern Jilin is still southerly. In Jilin, there is still a southwest-northeast convergence line, forming an inverted "human-shaped" convergence line. Heavy precipitation occurs in the northerly airflow in the lower reaches of the Nenjiang River Basin, and there is no ground lightning activity; at 21:00 on August 3, 2023, the inverted "human-shaped" convergence line still exists ( Fig.10 b) Heavy rainfall occurs on the north and east sides of the convergence line, and there is no ground lightning activity in the heavy rainfall area, which will last until 21:00 on August 3, 2023 ( Fig.10 c) There is no lightning activity in the heavy rainfall area; 00:00 on August 4, 2023 ( Fig.10 d) The inverted "Z-shaped" convergence lines merge into north-south convergence lines, there is no ground lightning, and the heavy rainfall process is coming to an end.

[0117] In summary, the water vapor came from the Bohai Sea from 24 hours before to when the heavy rainfall occurred. The wind field showed that the convergence line existed 24 hours before the heavy rainfall, and appeared in front of the source of the water vapor channel. The appearance of the convergence line is the triggering mechanism for the occurrence of heavy rainfall, which has a dynamic lifting effect on the water vapor. At the same time, the convergence line is also the convergence area of ​​low-level cold air and warm and humid air currents. Convection is easily stimulated near the convergence line, resulting in heavy rainfall weather.

[0118] The heavy rainfall nowcasting method described in the embodiment of the present invention determines whether there is water vapor source from the Bohai Sea and the Sea of ​​Japan transporting and establishing to the heavy rainfall area through wind direction, and at the same time, determines whether there is wind speed pulsation feature in the water vapor channel before the heavy rainfall occurs; and determines the ground convergence line, cyclone convergence center, secondary cold front, wind speed convergence, wind speed and wind direction convergence through the wind field, providing a trigger mechanism for the occurrence of heavy rainfall, wherein the characteristics of the convergence line are mainly manifested as stable and less moving, north-south swing, northeastward movement, southeastward movement, and in a few individual cases, the early convergence line and the later wind speed convergence, and the occurrence of heavy rainfall. Convergence lines appear, and in addition, there are characteristics of mutual transition, alternation or common appearance between convergence lines and secondary cold fronts or cyclone convergence centers. When the source of water vapor for heavy precipitation is determined, the heavy precipitation area appears near the ground convergence lines and cyclone convergence centers. Under the condition that the water vapor source and triggering mechanism of wind field analysis already exist, the heavy precipitation area is determined by using the area where the density of ground lightning changes from sparse to dense near the ground convergence lines, the wind speed changes from high to low or the wind direction changes from warm and humid air flow direction to cold air direction on the water vapor source channel near the ground convergence lines, so as to realize the nowcast of heavy precipitation.

[0119] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A heavy rainfall nowcasting method, characterized in that: include: Extract raw data, including hourly precipitation data, hourly 2-minute average wind data, and hourly lightning location data; The original data are preprocessed to obtain multi-source data files, including multi-factor mapping data files, ground full-factor mapping data files and lightning location data files; the multi-source data files are displayed using the Micaps platform, the wind field is analyzed to identify the water vapor source and convergence line, the spatial configuration picture set of heavy precipitation area, wind field and ground lightning is drawn, and a heavy precipitation data set of wind field characteristics and ground lightning characteristics is constructed; based on the multi-source data files, the water vapor source characteristics, wind speed pulsation characteristics and ground triggering mechanism characteristics from 24 hours before the occurrence of heavy precipitation to the end are identified, and the heavy precipitation area is determined by using the evolution characteristics of ground lightning density from sparse to dense near the ground convergence line, as well as the area where the wind speed changes from large to small or the wind direction changes from warm and humid air flow direction to cold air direction on the water vapor source channel near the ground convergence line.

2. The heavy rainfall nowcasting method according to claim 1, characterized in that: Program code is written in C# language to extract raw data from the meteorological big data cloud platform and store it in txt text format; the fields of hourly precipitation data include address, station name, station number, longitude, latitude, altitude, precipitation and precipitation quality control code; the fields of hourly 2-minute average wind data include address, station name, station number, longitude, latitude, altitude, 2-minute average wind direction, 2-minute average wind speed, 2-minute average wind direction quality control code and 2-minute average wind speed quality control code; the fields of hourly lightning location data include record identification, data source, data identification, storage time, receipt time, update time, data time, latitude, longitude, positioning error, return stroke peak current intensity, return stroke maximum steepness, positioning method, lightning geographical location information and correction report mark.

3. The heavy rainfall nowcasting method according to claim 1, characterized in that: The method for processing hourly precipitation data includes: using C# language to write program code to extract data that meets the heavy precipitation standard and has a precipitation quality control code of 0 from the hourly precipitation data, and convert it into diamond 34 data of the Micaps platform, that is, obtaining a multi-factor mapping data file; based on the quality-controlled hourly precipitation data, the heavy precipitation case data is extracted, and the time period of each heavy precipitation case is: from the first heavy precipitation time to before two or more consecutive non-precipitation times.

4. The heavy rainfall nowcasting method according to claim 3, characterized in that: The method for processing the hourly 2-minute average wind data includes: using C# language to write program code to extract the 2-minute average wind speed and wind direction data whose 2-minute average wind direction quality control code and 2-minute average wind speed quality control code are both 0 from the 2-minute average wind data, and convert them into diamond 1 type data of the Micaps platform, that is, obtaining a ground full-factor mapping data file; the time period of the ground full-factor mapping data file is from 24 hours before the heavy rainfall occurs to the end of the heavy rainfall.

5. The heavy rainfall nowcasting method according to claim 4, characterized in that: The method for processing hourly lightning location data is: use C# language to write program code to convert lightning location data into diamond41 type data of Micaps platform, that is, to obtain lightning location data file; the time period of lightning location data file is from 24 hours before heavy rainfall occurs to the end of heavy rainfall.

6. The heavy rainfall nowcasting method according to claim 5, characterized in that: The identification process of water vapor source characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case, calling the diamond 34 data, diamond 1 data and diamond 41 data from 24 hours before the heavy rainfall to the same time of the end on the Micaps platform; analyzing the water vapor source through the wind field wind speed characteristics, and filling the water vapor source wind direction and wind speed evolution characteristics from 24 hours before the heavy rainfall to the end into the corresponding case in the heavy rainfall data set table one by one.

7. The heavy rainfall nowcasting method according to claim 5, characterized in that: The identification process of wind speed pulsation characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case, calling the diamond 34 type data, diamond 1 type data, and diamond 41 type data from 24 hours before the heavy rainfall to the same time of the end on the Micaps platform to analyze the wind speed pulsation characteristics; filling the wind speed pulsation analysis results from 24 hours before the heavy rainfall to the end into the corresponding cases in the heavy rainfall data set table one by one.

8. The heavy rainfall nowcasting method according to claim 7, characterized in that: There are two types of wind speed pulsation characteristics. One is that the wind speed in the water vapor channel in the same area has increased compared with the previous period, and the other is that the wind speed in the water vapor channel has increased from the sea surface to the heavy rainfall area.

9. The heavy rainfall nowcasting method according to claim 5, characterized in that: The identification process of the ground triggering mechanism characteristics includes: according to the occurrence time of heavy rainfall in the heavy rainfall case, calling the diamond 34 type data, diamond 1 type data, and diamond 41 type data from 24 hours before the occurrence of heavy rainfall to the same time of the end on the Micaps platform; analyzing the position and movement characteristics of the convergence line from 24 hours before the occurrence of heavy rainfall to the end; filling the analysis results of the time and position evolution characteristics of the convergence line from 24 hours before the occurrence of heavy rainfall to the end into the corresponding cases in the heavy rainfall data set table one by one; analyzing the evolution characteristics and density of ground-to-ground lightning activity near the convergence line; filling the dense evolution characteristics of ground-to-ground lightning from 24 hours before the occurrence of heavy rainfall to the end, and the configuration relationship between the dense ground-to-ground lightning area and heavy rainfall into the corresponding cases in the heavy rainfall data set table one by one.

10. The heavy rainfall nowcasting method according to claim 5, characterized in that: According to the occurrence time of heavy rainfall in individual heavy rainfall cases, the Micaps platform calls diamond 34 data, diamond 1 data, and diamond 41 data from 24 hours before the heavy rainfall to the same time when it ends to draw a set of spatial configuration pictures of heavy rainfall areas, wind fields, and ground flashes.

Images