Ground-to-ground lightning forecasting method and system based on vaterite particle analysis

Abstract

The invention provides a cloud-to-ground lightning forecasting method and system based on vaterite particle analysis. The method comprises the following steps: constructing a meteorological acquisition system including a meteorological radar system; acquiring and extracting information of vaterite particles of the fluid by using a meteorological acquisition system; classifying the information of the vaterite particles of the fluid acquired by the meteorological acquisition system; and calculating the vaterite particle layer thickness of the fluid according to the classified vaterite particle information, and setting a threshold value as a forecast index of ground-to-ground lightning. According to the vaterite particle analysis-based ground-to-ground lightning forecasting method and system provided by the invention, the ground-to-ground lightning forecasting advance can be increased, the forecasting accuracy can be improved, and the method and the system have important practical significance in reducing the severe convection ground-to-ground lightning early warning empty report rate and increasing the early warning advance time quantum.

Description

Technical Field

[0001] This invention proposes a ground flash prediction method and system based on graupel analysis, which relates to the field of atmospheric science. Background Technology

[0002] As a major component of severe convective weather, lightning activity is characterized by its sudden occurrence, rapid development, difficulty in forecasting its generation and dissipation, and strong destructive potential. Within the entire lightning lifecycle, ground flash initiation is the moment when fluid first discharges to the ground. Although ground flash initiation is only a very short stage in the entire lightning process, its timing is the starting point for extrapolating the entire lightning process and determines the forecast of the potentially affected areas. Therefore, strengthening research on ground flash initiation forecasting is of significant practical importance for reducing the false alarm rate of severe convective ground flash warnings and increasing the lead time for warnings.

[0003] In the field of lightning nowcasting, the mainstream approach is to use observational data or physical quantities related to lightning occurrence (such as radar, lightning locators, and satellite data) combined with tracking, extrapolation, and deep learning methods to calculate the lightning location within the next 0-2 hours. Simultaneously, it incorporates gridded products for short-term and potential lightning convection forecasts to supplement and revise the extrapolated warning results. These methods include single-cell centroid tracking extrapolation, regional comprehensive extrapolation, optical flow methods, and temporal convolutional neural network-based extrapolation methods. These research findings, besides including nowcasting, also involve the forecasting and analysis of the initial stage of ground flashes, noting the impact of the initial flash time and location on the accuracy of subsequent extrapolation forecasts. For example, some scholars suggest that the initial detection of a radar echo with an intensity of 10 dBz near the freeze layer may be a characteristic of future lightning; and that lightning may occur when the reflectivity threshold of 40 dBz is reached on two consecutive volume scans at a temperature of -10℃. Several scholars have used radar products to conduct research on lightning initiation forecasting, mainly focusing on the relationship between ZDR, KDP and lightning initiation. However, the results of their research are not entirely consistent. For example, some scholars believe that before the first lightning strike, the ZDR in the convective cloud continuously decreases and a negative and low value region appears above the -10 to 0°C layer. After the first lightning strike, the low value region of ZDR in the upper part of the echo further expands. Other scholars have found that there are significant changes in KDP and ZDR in the upper part of the convective cloud before and after the lightning strike, with the upper-level ZDR changing from a negative value before the lightning strike to a positive value. Some scholars have concluded that lightning is concentrated in the region with ZDR above -0.4 to 1.4 dB.

[0004] Overall, existing research indicates that using tools, including radar, to identify and analyze the particle characteristics of fluids above the 0°C layer can effectively deepen our understanding of the occurrence and development of lightning, especially ground lightning, and provide necessary guidance for lightning weather warnings. However, because the occurrence and development of lightning are affected by factors such as topography and atmospheric cloud physics, they exhibit great uncertainty, and different conclusions may be drawn for different regions and different lightning research subjects. Summary of the Invention

[0005] In view of this, in order to fill the gaps and deficiencies in the existing technology, this invention proposes a ground lightning forecasting method and system based on graupel analysis, which can increase the lead time of ground lightning forecasts and improve forecast accuracy. It has important practical significance for reducing the false alarm rate of severe convective ground lightning warnings and increasing the lead time of warnings.

[0006] This invention proposes a ground flash prediction method and system based on graupel analysis, including the following:

[0007] This invention proposes a ground lightning prediction method based on graupel analysis, characterized in that the ground lightning prediction method based on graupel analysis includes the following:

[0008] Construct a meteorological data acquisition system, including a meteorological radar system.

[0009] Information on graupel particles in fluids is collected and extracted using a meteorological data acquisition system.

[0010] The information on graupel particles in the fluid collected by the meteorological data acquisition system is classified.

[0011] The thickness of the graupel layer on the fluid is calculated based on the information of the classified graupel particles, and a threshold is set as a forecast indicator for ground flashes.

[0012] Furthermore, the meteorological data acquisition system includes the following:

[0013] The aforementioned weather radar system operates with at least two radars working simultaneously.

[0014] The aforementioned weather radar system operates in the S-band.

[0015] The meteorological data acquisition system also includes a lightning location system, which comprises several lightning location instruments.

[0016] Furthermore, the method of using a meteorological data acquisition system to collect and extract information on graupel particles from fluids includes the following:

[0017] The meteorological acquisition system collects information on graupel particles in the fluid, including: the combined radar reflectivity of the fluid in each scanning cycle and the particle phase classification information of each particle in the fluid corresponding to at least 9 different radar elevation angles in each scanning cycle.

[0018] Furthermore, the collection and extraction of information on graupel particles from fluids using meteorological data acquisition systems also includes the following:

[0019] Among the particle phase state classification information of each particle in the fluid, the particle phase state classification information that conforms to graupel is selected. The information of graupel in the fluid includes the reflectivity factor of graupel in the fluid, the differential reflectivity of graupel, the cocorrelation coefficient of graupel, the differential propagation phase shift of graupel, the specific differential phase shift of graupel, and the backscattering phase of graupel.

[0020] Furthermore, the classification of information on graupel particles in the fluid collected by the meteorological data acquisition system includes the following:

[0021] The graupel layer thickness data includes both positive and negative sample data;

[0022] Among them, the graupel layer thickness data when there is a ground flash during the fluid's life cycle is recorded as positive sample data, and the data when there is no ground flash during the fluid's life cycle is recorded as negative sample data.

[0023] When a ground flash occurs during the life cycle of a convective fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the positive sample radar, taking the time point of the first ground flash as the starting point.

[0024] The closed area with radar combined reflectivity CR≥40dBZ and covering the location of ground flash during the working time of the positive sample radar is defined as the positive sample area.

[0025] Furthermore, the graupel layer thickness of all fluid pairs within the positive sample region was statistically analyzed within 60 minutes before the occurrence of ground flash. For fluid pairs with graupel layer thickness data of less than 60 minutes, only the time period with graupel layer thickness data was taken; for fluid pairs with graupel layer thickness data of more than 60 minutes, only the data within 60 minutes was taken. Finally, the positive sample data was obtained.

[0026] Furthermore, the classification of information on graupel particles in the fluid obtained by the meteorological data acquisition system also includes the following:

[0027] When no ground flash occurs during the life cycle of the fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the negative sample radar, starting from the moment when the radar combined reflectivity CR≥40dBZ of the fluid first appears.

[0028] The moment when the radar combined reflectivity CR ≥ 40dBZ last appeared in the fluid during the radar's operating time was taken as the endpoint. The closed area with radar combined reflectivity CR ≥ 40dBZ within the start and end time period was selected as the negative sample area. The thickness of the graupel layer in the negative sample area was statistically analyzed.

[0029] Furthermore, the calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flash includes the following:

[0030] The heights corresponding to the radar elevation angles of all locations where graupels appear are stored in the database. Finally, the difference between the maximum and minimum heights in the database is calculated and used as a threshold. When the thickness of the graupel layer of a new graupel exceeds the threshold, it is determined that a ground flash will occur in the area where the new graupel appears, and a forecast is made.

[0031] Furthermore, the calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flash also includes the following:

[0032] When the graupel layer thickness of a new fluid exceeds a threshold, it is predicted that a ground flash will occur in that new fluid when more than 95% of the graupel layer thickness samples at all sampling times are greater than the threshold.

[0033] According to a second aspect of the present invention, a ground flash prediction system based on graupel analysis is proposed, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements a ground flash prediction method based on graupel analysis as described in any of the present invention.

[0034] According to a third aspect of the present invention, a ground flash prediction system based on graupel analysis is proposed, comprising a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements a ground flash prediction method based on graupel analysis as described in any of the present invention.

[0035] The present invention has the following advantages:

[0036] This invention proposes a ground lightning forecasting method and system based on graupel analysis, which can increase the forecast lead time and improve the forecast accuracy. This has significant practical implications for reducing the false alarm rate of severe convective ground lightning warnings and increasing the warning lead time. This invention improves the accuracy of graupel layer thickness calculation. By integrating data from multiple radars, it overcomes the observation bias caused by single radar limitations such as detection angle restrictions and obstructions. This invention expands the observation range. The combined use of multiple radars can cover a larger area and improve the ability to monitor fluids at long distances. This invention enhances system reliability. The complementarity of data from multiple radars can reduce the risk of data loss due to single radar failure or maintenance. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the present invention.

[0038] Figure 2 This is a schematic diagram of the detection range of the meteorological data acquisition system of the present invention.

[0039] Figure 3 This is a schematic diagram of the thickness of the graupel layer in this invention.

[0040] Figure 4 This is a schematic diagram of the time-series quantile value curve of the thickness of the graupel layer in the fluid with and without ground flash, according to the present invention.

[0041] Figure 5 This is a schematic diagram of the time sequence color blocks for the thickness of the graupel layer in a fluid with or without ground flash, according to the present invention.

[0042] Figure 6 This is a schematic diagram of the statistical box plot of the thickness of the initial graupel layer in lightning events according to the present invention.

[0043] Figure 7 This is a schematic diagram of the combined reflectance of four fluid pairs according to an embodiment of the present invention.

[0044] Figure 8 This is a schematic diagram of the temporal distribution of the graupel layer thickness according to an embodiment of the present invention.

[0045] Figure 9 This is a schematic diagram of the timing distribution of radar combined reflectivity according to an embodiment of the present invention.

[0046] Figure 10 This is a schematic diagram of the time-series distribution of vertical liquid water content according to an embodiment of the present invention.

[0047] Figure 11 This is a schematic diagram of the echo top height timing distribution according to an embodiment of the present invention.

[0048] Figure 12 This is a schematic diagram comparing the time sequence of fluid graupel layer height and lightning frequency for T1 and T2 in an embodiment of the present invention. Detailed Implementation

[0049] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0050] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0051] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention; as used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise; furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0052] like Figure 1 As shown, this invention proposes a ground flash prediction method based on graupel analysis, characterized by the following:

[0053] Construct a meteorological data acquisition system, including a meteorological radar system.

[0054] Information on graupel particles in fluids is collected and extracted using a meteorological data acquisition system.

[0055] The information on graupel particles in the fluid collected by the meteorological data acquisition system is classified.

[0056] The thickness of the graupel layer on the fluid is calculated based on the information of the classified graupel particles, and a threshold is set as a forecast indicator for ground flashes.

[0057] The meteorological data acquisition system includes the following:

[0058] The aforementioned weather radar system operates with at least two radars working simultaneously.

[0059] The aforementioned weather radar system operates in the S-band.

[0060] The meteorological data acquisition system also includes a lightning location system, which comprises several lightning location instruments.

[0061] Furthermore, in one embodiment of the present invention, the weather radar system employs at least two S-band fully coherent dual-polarization Doppler weather radars. The scanning cycle of the S-band fully coherent dual-polarization Doppler weather radar is one volume scan every 6 minutes.

[0062] Furthermore, in one embodiment of the present invention, the lightning location system uses ground flash data from the ADTD lightning location system, which consists of 16 lightning location devices.

[0063] Furthermore, in one embodiment of the present invention, since the ground flash location data of the ADTD lightning location system has a certain probability of error, in order to avoid interference from erroneous ground flashes on the selection and analysis of research objects, quality control based on radar combined reflectivity is performed on the ground flash data before case selection and analysis. Therefore, the present invention selects 25 dBz as the benchmark for ground flash-lightning combined reflectivity quality control, using 0.01° × 0.01° as the basic grid. If the radar combined reflectivity CR of a certain ground flash location latitude and longitude grid point and its surrounding one grid point (a total of 9 grid points) does not exceed 25 dBz at the radar time closest to the ground flash time and the two times before and after it, the ground flash is discarded. The final effect is as follows: Figure 2 As shown.

[0064] The aforementioned method of collecting and extracting information on graupel particles from fluids using a meteorological data acquisition system includes the following:

[0065] The meteorological acquisition system collects information on graupel particles in the fluid, including: the combined radar reflectivity of the fluid in each scanning cycle and the particle phase classification information of each particle in the fluid corresponding to at least 9 different radar elevation angles in each scanning cycle.

[0066] Among them, in the particle phase state classification information of each particle in the fluid, the particle phase state classification information that conforms to the graupel particle is selected. The information of graupel particles in the fluid includes the reflectivity factor of graupel particles in the fluid, the differential reflectivity of graupel particles, the cocorrelation coefficient of graupel particles, the differential propagation phase shift of graupel particles, the specific differential phase shift of graupel particles, and the backscattering phase of graupel particles.

[0067] Furthermore, in one embodiment of the present invention, the fluid is selected to always be within the 200km detection range of at least two radars throughout its life cycle (e.g., Figure 2 The fluid within the area shown is taken as the research object. It is required that the fluid trajectory exists continuously within this area. The corresponding reflectivity CR and the particle phase classification HCL product at nine elevation angles (0.5°, 1.5°, 2.4°, 3.4°, 4.3°, 6.0°, 9.9°, 14.6°, 19.5°) are extracted. The particle phase classification product uses a fuzzy logic algorithm (Hyangsuk Park, 2009) to distinguish six types of data including Z, ZDR, V, CC, PDP, and KDP, and then obtains ten particle classification results including light rain, heavy rain, hail, large raindrops, clear sky echo, ground objects, dry snow, wet snow, ice crystals, and graupel.

[0068] Furthermore, in one embodiment of the present invention, in order to simultaneously take into account the characteristics of fluids in which no ground flash occurs, a value slightly lower than that of existing research results (40 dBZ) is selected as the standard to divide the fluid study areas of positive and negative samples.

[0069] Furthermore, the classification of information on graupel particles in the fluid collected by the meteorological data acquisition system includes the following:

[0070] The graupel layer thickness data includes both positive and negative sample data;

[0071] Among them, the graupel layer thickness data when there is a ground flash during the fluid's life cycle is recorded as positive sample data, and the data when there is no ground flash during the fluid's life cycle is recorded as negative sample data.

[0072] When a ground flash occurs during the life cycle of a convective fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the positive sample radar, taking the time point of the first ground flash as the starting point.

[0073] The closed area with radar combined reflectivity CR≥40dBZ and covering the location of ground flash during the working time of the positive sample radar is defined as the positive sample area.

[0074] Furthermore, the graupel layer thickness of all fluid pairs within the positive sample region was statistically analyzed within 60 minutes before the occurrence of ground flash. For fluid pairs with graupel layer thickness data of less than 60 minutes, only the time period with graupel layer thickness data was taken; for fluid pairs with graupel layer thickness data of more than 60 minutes, only the data within 60 minutes was taken. Finally, the positive sample data was obtained.

[0075] When no ground flash occurs during the life cycle of the fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the negative sample radar, starting from the moment when the radar combined reflectivity CR≥40dBZ of the fluid first appears.

[0076] The moment when the radar combined reflectivity CR ≥ 40dBZ last appeared in the fluid during the radar's operating time was taken as the endpoint. The closed area with radar combined reflectivity CR ≥ 40dBZ within the start and end time period was selected as the negative sample area. The thickness of the graupel layer in the negative sample area was statistically analyzed.

[0077] Furthermore, the calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flash includes the following:

[0078] The heights corresponding to the radar elevation angles of all locations where graupels appear are stored in the database. Finally, the difference between the maximum and minimum heights in the database is calculated and used as a threshold. When the thickness of the graupel layer of a new graupel exceeds the threshold, it is determined that a ground flash will occur in the area where the new graupel appears, and a forecast is made.

[0079] Furthermore, in one embodiment of the present invention, the height of the lowest elevation angle at which graupel appears in the fluid is denoted as h1, and the height in the highest elevation angle diagram where graupel exists is denoted as h2. h2-h1 represents the thickness of the graupel layer for a single fluid instance. Since the graupel products of the S-band fully coherent dual-polarization Doppler weather radar are in radial data format, the graupel layer thickness is calculated based on the vertical distance. It is necessary to analyze the graupel distribution at different elevation angles at the same latitude and longitude location before performing the calculation. Due to limitations in detection angle, obstruction, and other factors, the calculation of the graupel layer thickness by a single radar may have a significant deviation. Figure 3 As shown, the optimal calculation result is obtained by fusing the angle1 of radar 1 and the angle2 of radar 2. Therefore, the actual thickness of the graupel layer is calculated by integrating the data from multiple radars within the fluid range. The specific calculation method is to iterate through all elevation angles of each radar at each moment, store the heights corresponding to all elevation angles where graupel appears at the fluid location in a list, and finally calculate the difference between the maximum and minimum values ​​in the list. This is the graupel layer thickness value of the fluid obtained by integrating the data from multiple radars. The graupel height is calculated using the radar elevation angle.

[0080] The calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flashes also includes the following:

[0081] When the graupel layer thickness of a new fluid exceeds a threshold, it is predicted that a ground flash will occur in that new fluid when more than 95% of the graupel layer thickness samples at all sampling times are greater than the threshold.

[0082] Furthermore, in one embodiment of the present invention, based on the method of the present invention, 32 convective process samples with and without ground flashes were obtained in a certain region over a period of 2 years. The thickness of the graupel layer was used as the ordinate, with the moment of ground flash initiation for positive samples and the moment when the combined reflectance CR of the convective fluid finally reached 40 dBZ for negative samples as the endpoints of the abscissa. For all samples at each abscissa moment, the 95th percentile, 75th percentile, median, 25th percentile, and 5th percentile were calculated. All samples with the same moment were connected to form a curve, as shown below. Figure 4 The distribution diagram is shown below; for the 32 positive samples and 22 negative samples, different colors are filled according to the grenade layer thickness value of each sample and each time step to draw a sample time-series color block diagram, as shown. Figure 5 As shown.

[0083] Furthermore, in one embodiment of the present invention, analysis Figure 4 It can be seen that in fluid samples with initial lightning strikes, the graupel layer thickness shows a gradually increasing trend, with all samples reaching a graupel layer thickness of at least 2.26 km at the moment of the first lightning strike (the minimum value of the sample at time 0). In fluid samples without lightning strikes, the graupel layer thickness shows a trend of first increasing and then decreasing with the life cycle of the fluid, with the graupel layer thickness being less than 2.2 km at most times (95th percentile), and the maximum graupel layer thickness only reaching 2.5 km. Figure 4 Comparing the positive and negative datasets reveals that the color values ​​of the positive sample color patches are significantly greater than those of the negative sample color patches. The average values ​​for the positive and negative datasets are 2.38 km and 1.20 km, respectively. Since the graupel layer thickness in both positive and negative samples shows an increasing trend during fluid development, and there are certain differences in the numerical range of graupel layer thickness between the positive and negative samples, a certain graupel layer thickness threshold can be extracted to identify the probability that any sample belongs to either the positive or negative sample. By combining the time lead of the extracted graupel layer thickness threshold in the positive sample, the prediction of the initiation of ground flash in the fluid at a future moment can be achieved.

[0084] Figure 4 , Figure 5 The distribution maps of graupel thickness layers for all positive and negative samples in chronological order are given. Figure 4 , Figure 5 It can be seen that in fluid samples where ground flash occurred (initial formation), the graupel layer thickness was greater than 2 km in over 95% of the samples within the -12 min to -18 min time interval. Figure 4 (Blue line) In the 22 convective samples without ground flashes, only 5 samples (5 / 22 = 22.72%) showed a graupel layer thickness greater than 2 km during convection development. Based on the above forecast indicators, the forecast effectiveness of the total 32+22 samples was analyzed, and the results are shown in Table 1. The calculated TS = 0.864. The advance forecast time of all positive samples under the above forecast indicators was summarized, with the minimum advance forecast being 12 min, the maximum advance forecast being 48 min, and the average advance forecast being 28.13 min. Without considering other meteorological data, relying solely on graupel layer thickness to identify the initial formation of ground flashes has certain forecasting potential.

[0085] The sample prediction results under the condition that the graupel layer exceeds 2km are shown in Table 1:

[0086]

[0087] Table 1. Sample forecast results under the condition that the graupel layer exceeds 2 km.

[0088] The graupel layer thickness in all individual cases of convective weather was statistically analyzed within 66 minutes prior to the first lightning strike. Based on the background characteristics of lightning weather in a given region, the samples were categorized into summer subtropical high-controlled or marginal types, cold-warm shear lines, and trough activity types. Box plots and median line plots were generated according to the weather classification criteria selected for each case (e.g., Figure 6 As shown in the figure, the initial formation of lightning along the cold and warm shear lines and troughs requires a longer charge accumulation time. The appearance and thickening of the graupel layer takes 1-2 body scan times longer than that of lightning controlled or bordered by the subtropical high in summer. The thickness and dispersion of the graupel layer at the time of initial formation are slightly larger than those of lightning controlled or bordered by the subtropical high. In contrast, lightning controlled or bordered by the subtropical high does not require a longer accumulation time before initial formation, and the initial formation speed of lightning is relatively fast. In some cases, graupel was observed at both elevation angles from the beginning, and the thickness exceeded 2 km, after only 3-4 lightning strikes. The first ground flash occurs quickly upon arrival, and the data deviation of individual samples is relatively small. In addition, cold and warm shear lines are more likely to produce scattered graupel, which may be related to the strength of the background dynamics, thermal and water vapor of the convective fluid. In contrast, graupel is mostly distributed in patches in the convective fluid at the edge of the subtropical high. However, regardless of the weather type, the temporal characteristics of the graupel layer thickness are not different from the relationship between whether ground flash occurs (initial formation) in the convective fluid. Different weather types have no significant impact on the use of graupel layer thickness as a ground flash forecast indicator.

[0089] The charging between graupel and ice crystal particles is the main way to accumulate charge in the early stages of lightning. Therefore, when graupel appears, charge begins to accumulate rapidly in the fluid. The generation of lightning requires sufficient charge to generate an electric field that can break down the air. The continuous and abundant presence of graupel helps to meet this condition. This is reflected in the characteristics of dual-polarization HCL graupel identification products, which means that after graupel appears, it must persist for a period of time and reach a certain thickness before it is possible to accumulate enough charge to generate and sustain lightning. In negative samples (fluids where no ground flash occurred), the thickness of the graupel layer is much smaller than that in positive samples, and the fluid does not accumulate enough charge, so no lightning ultimately occurs.

[0090] Based on the above analysis, "graupel layer thickness exceeding the 2km threshold" has certain indicative significance for whether ground flash will occur in a certain region in the future, and can be used as a major indicator for predicting the initial formation of ground flash in graupel.

[0091] According to a second aspect of the present invention, a ground flash prediction system based on graupel analysis is proposed, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, implements a ground flash prediction method based on graupel analysis as described in any of the present invention.

[0092] According to a third aspect of the present invention, a ground flash prediction system based on graupel analysis is proposed, comprising a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements a ground flash prediction method based on graupel analysis as described in any of the present invention.

[0093] In addition to the above, the present invention also has related embodiments, including the following:

[0094] In one embodiment of the present invention, from 14:00 to 16:00 on a certain day, month, and year, a low-level shear line affected a certain region from west to east. The region had good water vapor conditions, with a K-index above 35°C, indicating strong convective potential. Around 14:15, convective clouds triggered by the shear line formed, gradually developing eastward and entering the detection range of the Xiamen dual-polarization radar. From 14:15 to 16:10, the convective clouds moved out of the radar's detection range. Convective clouds within the region, labeled T1, T2, F1, and F2 (e.g.,...), were observed. Figure 7 As shown), radar time-series tracking was performed to calculate and plot the variation and time-series curves of the fluid graupel layer thickness for each pair (as shown). Figure 8 ), where T1 and T2 are the convective fluids where future lightning strikes (initial ones) are possible, and F1 and F2 are the convective fluids where future lightning strikes are not possible. Simultaneously, ADTD lightning location data is read, and quality control is performed according to the data processing method of this invention. The presence of lightning strikes at corresponding times is statistically analyzed. Figure 8 The corresponding curve is marked with a "glowing" indicator.

[0095] Furthermore, in one embodiment of the present invention, by Figure 8It can be seen that for fluid T1, from the moment it entered the radar detection area until it left, graupel was observed at at least two elevation angles. The graupel layer thickness exceeded 2 km (2.36 km) at 14:39, and the first ground flash occurred at 14:45. As time progressed, the graupel layer thickness showed an increasing trend, reaching a maximum of 5.92 km at 15:37. Subsequently, the graupel layer thickness began to decrease, while ground flash activity continued until the fluid moved out of the radar detection range. For fluid T2, graupel activity was observed simultaneously at two elevation angles starting at 14:28, but the graupel layer thickness was less than 2 km for the next hour until it exceeded 2 km (2.39 km) at 15:26. The first ground flash occurred at 15:31, at which time the graupel layer thickness reached its maximum. Subsequently, the graupel layer thickness began to decrease, and ground flash activity continued until 15:49 (a total of 24 minutes). At 15:08 and 14:45, respectively, graupel activity was simultaneously observed in fluids F1 and F2 at two or more elevation angles. However, until F1 and F2 moved out of the radar detection range, the graupel layer thickness of neither exceeded 2 km, and neither of them ultimately experienced ground flash. The ground flash initiation index of graupel layer thickness exceeding 2 km obtained according to the embodiment of the present invention was used to predict the ground flash initiation of four fluid pairs. The results showed that all four fluid pairs provided accurate qualitative predictions of whether or not a ground flash (initiation) occurred. However, the initiation prediction lead time for fluids T1 and T2 was relatively small, with both providing ground flash initiation predictions only 6 minutes in advance.

[0096] Furthermore, in one embodiment of the present invention, the radar combined reflectivity, liquid water content, and maximum echo top height are extracted for each fluid pair at each time step, and the corresponding time-series curves are plotted, as shown in the figure. Figure 9 , Figure 10 , Figure 11 As shown, by Figure 9 It can be seen that there is no significant difference in the combined reflectivity of the four pairs of fluids, and during the peak development period, their combined reflectivity all reached over 55 dBZ. Figure 10 , Figure 11The results show that the liquid water content and echo top height of fluid T1 are greater than those of the other three fluid pairs. The average echo top height is 11.63 km, and the maximum value is close to 15.3 km. The relationship and trend with the graupel layer thickness of T1 are relatively consistent. The differences in liquid water content and echo top height between T2, F1 and F2 are weak, and the echo top height is maintained at around 8 km (average 8.095). Regarding the temporal distribution characteristics of graupel layer thickness, the values ​​for T1 and T2, the two pairs of fluids with ground flashes, are relatively high, with average values ​​of 2.75 km and 1.87 km, and maximum values ​​of 5.92 km and 3.5 km, respectively. In contrast, the average values ​​for F1 and F2, which did not experience ground flashes, are 1.54 km and 1.34 km, with maximum values ​​of 1.98 km and 1.81 km, respectively. Calculating the difference in graupel layer thickness (the value at one time step minus the value at the previous time step, i.e., the change between adjacent time steps) for the four pairs of fluids reveals that the percentages of times with a change greater than 0 for T1, T2, F1, and F2 are 63%, 67%, 62%, and 47%, respectively. The graupel layer thickness for the pairs of fluids with ground flashes is continuously increasing over more time steps, and for T1 and T2, the rate of change of graupel layer thickness is increasing in the few time steps before the first ground flash. Therefore, in this specific case, graupel layer thickness is more significant in characterizing ground flash activity compared to other common radar products.

[0097] Furthermore, in one embodiment of the present invention, a comparison graph of the height of the fluid graupel layer and the number of lightning strikes for T1 and T2 is plotted (e.g., Figure 12 ),Depend on Figure 12 It can be seen that the value of the graupel thickness layer shows a certain correlation with the number of ground flashes, which is consistent with the present invention. Figure 6 The trends are quite similar; the thickness of the graupel layer in T1 is greater than that in T2 for most of the time. Lightning activity occurred continuously in T1 starting at 14:48, while lightning activity only occurred in the 24 minutes after 15:30 in T2, and the number of lightning strikes was also less than that in T1. The duration and frequency of lightning strikes per unit time may also be related to the thickness of the graupel layer. When the thickness of the graupel layer exceeds the peak and begins to decrease, the lightning process continues. This may be because the amount of charge and electric field accumulated in the fluid can still maintain lightning activity, so the phenomenon of lightning continuing even when the thickness of the graupel layer decreases occurs. However, as the thickness of the graupel layer decreases, the lightning activity gradually weakens and disappears.

[0098] In summary, this invention uses dual-polarization radar HCL products from a specific region on a specific date and time, along with ADTD lightning location data (ground flash) for that region, as data sources. It selects 32 convective processes with ground flash occurrences and 22 without ground flash occurrences as samples. Statistical analysis is performed on the temporal characteristics of the dual-polarization radar HCL graupel layer thickness during the 60-minute period before the first ground flash in convective processes with ground flash occurrences, and during the time period when the maximum combined reflectivity of convective processes without ground flash occurrences is greater than 40 dBZ. The main conclusions are as follows:

[0099] Compared to radar composite reflectivity, vertical liquid water content, and echo top height, graupel layer thickness is a more significant temporal characteristic in characterizing whether convective fluids will exhibit ground flash activity. Analysis shows that "graupel layer thickness exceeding 2 km" can effectively distinguish whether convective fluids in a certain region will exhibit ground flash activity during their development. The TS score for 54 samples was 0.864, with a minimum forecast advance of 12 minutes, a maximum forecast advance of 48 minutes, and an average forecast advance of 28.13 minutes.

[0100] This invention is not limited to the preferred embodiments described above. The specific meteorological parameters appearing in this invention are exemplary data and do not constitute a limitation thereof.

[0101] The present invention can also be implemented in the form of software, hardware, or a combination of software and hardware; wherein the software can be stored in a computer-readable medium and executed by a processor to perform the corresponding function.

[0102] Suitable computer-readable media include, but are not limited to, hard disks, flash memory, read-only memory (ROM), random access memory (RAM), and other media capable of storing program code.

[0103] The execution order of the steps described in the flowchart or logic block diagram can be adjusted or parallelized as needed, provided that it does not affect the implementation of the function.

[0104] The accompanying drawings referenced in this specification are for illustrative purposes only. Their dimensions, scale, or colors may be adjusted according to actual production needs. The reference numerals in the drawings should not be construed as limiting the scope of protection.

[0105] The above are preferred embodiments of the present invention. Any changes made to the technical solution of the present invention that do not exceed the scope of the technical solution of the present invention shall fall within the protection scope of the present invention.

Claims

1. A ground lightning prediction method based on graupel analysis, characterized in that, The aforementioned ground flash prediction method based on graupel analysis includes the following: Construct a meteorological data acquisition system, including a meteorological radar system; Information on graupel particles in fluids is collected and extracted using a meteorological data acquisition system. The information on graupel particles in the fluid collected by the meteorological data acquisition system is classified. The thickness of the graupel layer on the fluid is calculated based on the information of the classified graupel particles, and a threshold is set as a forecast indicator for ground flashes.

2. The ground flash prediction method based on graupel analysis according to claim 1, characterized in that, The meteorological data acquisition system includes the following: The aforementioned weather radar system operates with at least two radars working simultaneously. The aforementioned weather radar system operates in the S-band. The meteorological data acquisition system also includes a lightning location system, which comprises several lightning location instruments.

3. The ground flash prediction method based on graupel analysis according to claim 1, characterized in that, The aforementioned method of collecting and extracting information on graupel particles from fluids using a meteorological data acquisition system includes the following: The meteorological acquisition system collects information on graupel particles in the fluid, including: the combined radar reflectivity of the fluid in each scanning cycle and the particle phase classification information of each particle in the fluid corresponding to at least 9 different radar elevation angles in each scanning cycle.

4. The ground flash prediction method based on graupel analysis according to claim 3, characterized in that, The collection and extraction of information on graupel particles from fluids using meteorological data acquisition systems also includes the following: Among the particle phase state classification information of each particle in the fluid, the particle phase state classification information that conforms to graupel is selected. The information of graupel in the fluid includes the reflectivity factor of graupel in the fluid, the differential reflectivity of graupel, the cocorrelation coefficient of graupel, the differential propagation phase shift of graupel, the specific differential phase shift of graupel, and the backscattering phase of graupel.

5. The ground flash prediction method based on graupel analysis according to claim 1, characterized in that, The classification of information on graupel particles in fluids collected by the meteorological data acquisition system includes the following: The graupel layer thickness data includes both positive and negative sample data; Among them, the graupel layer thickness data when there is a ground flash during the fluid's life cycle is recorded as positive sample data, and the data when there is no ground flash during the fluid's life cycle is recorded as negative sample data. When a ground flash occurs during the life cycle of a convective fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the positive sample radar, taking the time point of the first ground flash as the starting point. The closed area with radar combined reflectivity CR≥40dBZ and covering the location of ground flash during the working time of the positive sample radar is defined as the positive sample area. Furthermore, the graupel layer thickness of all fluid pairs within the positive sample region was statistically analyzed within 60 minutes before the occurrence of ground flash. For fluid pairs with graupel layer thickness data of less than 60 minutes, only the time period with graupel layer thickness data was taken; for fluid pairs with graupel layer thickness data of more than 60 minutes, only the data within 60 minutes was taken. Finally, the positive sample data was obtained.

6. The ground flash prediction method based on graupel analysis according to claim 5, characterized in that, The classification of information on graupel particles in fluids acquired by the meteorological data acquisition system also includes the following: When no ground flash occurs during the life cycle of the fluid, the radar in the meteorological acquisition system that is closest to the first ground flash is selected as the negative sample radar, starting from the moment when the radar combined reflectivity CR≥40dBZ of the fluid first appears. The moment when the radar combined reflectivity CR ≥ 40dBZ last appeared in the fluid during the radar's operating time was taken as the endpoint. The closed area with radar combined reflectivity CR ≥ 40dBZ within the start and end time period was selected as the negative sample area. The thickness of the graupel layer in the negative sample area was statistically analyzed.

7. The ground flash prediction method based on graupel analysis according to claim 1, characterized in that, The calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flash includes the following: The heights corresponding to the radar elevation angles of all locations where graupels appear are stored in the database. Finally, the difference between the maximum and minimum heights in the database is calculated and used as a threshold. When the thickness of the graupel layer of a new graupel exceeds the threshold, it is determined that a ground flash will occur in the area where the new graupel appears, and a forecast is made.

8. The ground flash prediction method based on graupel analysis according to claim 7, characterized in that, The calculation of the graupel layer thickness of the fluid and the setting of a threshold as a prediction indicator for ground flash also includes the following: When the graupel layer thickness of a new fluid exceeds a threshold, it is predicted that a ground flash will occur in that new fluid when more than 95% of the graupel layer thickness samples at all sampling times are greater than the threshold.

9. A ground lightning prediction system based on graupel analysis, comprising an electronic device, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a ground flash prediction method based on graupel analysis as described in any one of claims 1 to 8.

10. A ground flash prediction system based on graupel analysis, comprising a computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a ground flash prediction method based on graupel analysis as described in any one of claims 1 to 8.

Images