Hail refined risk assessment method based on Doppler weather radar data

Abstract

The invention discloses a hail refined risk assessment method based on Doppler weather radar data, and relates to the technical field of weather radar data analysis and risk assessment, and the technical key points are as follows: S1, data collection and preprocessing; s2, radar data time downscaling processing; s3, establishing a refined hail process strength historical data set; s4, calculating disaster damage rates corresponding to different disaster intensity grades; s5, refining an intensity index historical data set in the hail process; s6, establishing a hail disaster risk assessment model; s7, assessing the exposure degree of the disaster-bearing body; and S8, carrying out refined risk assessment and zoning on hail disasters. Starting from the aspects of radar data processing, hail identification indexes, time accumulation effects and the like, a traditional hail disaster assessment method is improved by using radar data, the spatial resolution of an obtained hail intensity assessment result is higher, and a refined risk assessment result can provide a scientific decision basis for government disaster prevention and reduction work.

Description

Technical Field

[0001] The present invention relates to the technical field of weather radar data analysis, and particularly relates to a method for fine-grained hail risk assessment based on Doppler weather radar data. Background Art

[0002] Hail disaster assessment is of great significance to many fields such as agriculture, insurance, and infrastructure. At present, the current hail disaster risk assessment work mainly relies on conventional observation station data during the hail hazard assessment process. Some studies have also introduced hailfall data from artificial hail prevention operation points in the past ten years or so. The station observation data all have the disadvantages of sparse distribution and taking points to represent the whole. The hailfall records of artificial hail prevention operation points also have problems such as uneven distribution of stations, incomplete data length levels, and non-standard records.

[0003] Hail intensity assessment divides hail grades according to the hail diameter. The intensity of a single hail process not only depends on the distribution of hail diameters, but also needs to comprehensively consider factors such as the duration of hailfall and the accumulation of hail. At present, the assessment of the disaster losses of a single hail process mainly relies on remote sensing technology means. For example, a large number of achievements in the past two years have shown that the potential of object-based image analysis using unmanned aerial vehicle (UAV) aerial images and high-resolution satellite data in assessing crop hail losses is great. However, the cloud occlusion problem of high-resolution satellite data in mountainous areas leads to low coverage of disaster data, and the timeliness is inferior to that of weather radar. Using UAV image data for assessment after a disaster process has high costs, poor timeliness, and great difficulty in operationalization. Weather radar has the advantages of wide coverage and high observation accuracy. In assessing the intensity of regional hail processes, fine-grained process assessment results can be obtained according to the hailfall area identification method and time integral processing. Summary of the Invention

[0004] The purpose of the present invention is to solve the above problems and provide a method for fine-grained hail risk assessment based on Doppler weather radar data.

[0005] In order to achieve the above purpose, the technical solution of the present invention is as follows: The present invention provides a method for fine-grained hail risk assessment based on Doppler weather radar data, including the following steps: Step S1, data collection and preprocessing: Collect radar data, sounding data, hailfall data, and hail disaster loss data, and divide the hailfall data into three intensity grades according to the hail grade division method; Step S2, radar data preprocessing: Determine the time window of the radar data used for calculation according to the time records of historical hailfall events, and perform interpolation processing on the data of two adjacent time instances to obtain data with a time resolution of 1 minute; Step S3. Establish a refined historical dataset of hail process intensity: According to the hail identification index, the expected maximum hail diameter algorithm, and the time integration method, using the minute-by-minute radar three-dimensional reflectivity factor data, calculate the duration and maximum hail diameter of the hail process in the study area, and obtain a refined dataset of the hail process intensity distribution in the study area according to the hail grade classification method; Step S4. Calculate the disaster loss rates corresponding to different disaster intensity levels: Select the evaluation factors for hail disaster losses according to the evaluation object, and combine the total value of the evaluation factors in the year when the disaster occurred to calculate the disaster loss rates of each historical disaster process. According to the probability density distribution function of the disaster loss rate, use the standard deviation method to divide the disaster loss rates corresponding to different hail disaster intensities; Step S5. Refined intensity index historical dataset of hail process: Screen out the time periods that meet the hail identification index within the time window, calculate the cumulative hail time and the distribution of the maximum hail diameter over time during this time period, use the time-weighted integration method to calculate the refined hail intensity index S, and calculate the intensity index distribution of each hail process in the study area to form a refined intensity index historical dataset of hail process; Step S6. Hail disaster risk assessment: The hail disaster risk index H is calculated according to the following formula:

[0006] where H is the process intensity index evaluated by radar data, n represents all the time instances within the hail process time window, and the frequencies of each disaster-causing factor reaching the thresholds of mild, moderate, and severe disaster-causing factors are , and the disaster loss rates of their mild, moderate, and severe disasters are respectively ; Step S7. Assessment of the exposure of disaster-bearing bodies: The total amount or value within the unit area where the evaluation object is exposed to the hail disaster risk is used as the exposure assessment index (E) after normalization processing; Step S8. Refined risk assessment and zoning of hail disasters: Considering the hail disaster risk and the exposure of disaster-bearing bodies, construct an evaluation model for the refined risk index R of hail disasters;

[0007] In the formula, R is the refined risk index of hail disasters, h and e are weight coefficients, h + e = 1, which are determined according to the analytic hierarchy process or the expert scoring method. The hail disaster risk index is divided into five levels according to the standard deviation method, corresponding to high, relatively high, medium, relatively low, and low risk levels respectively.

[0008] The present invention is further configured according to "Hail Grades (GBT 27957-2011)": the hail grades are divided into small hail, medium hail, large hail and extra-large hail according to the hail diameter: the maximum diameter of small hail is less than 5 mm; the maximum diameter of medium hail is between 5 and 20 mm, the maximum diameter of large hail is between 20 and 50 mm, and the diameter of extra-large hail is greater than 50 mm. The large hail and extra-large hail are classified as the severe grade, the small hail corresponds to the mild grade hail, and the medium hail corresponds to the moderate grade hail.

[0009] The present invention is further configured that the specific sub-steps of step S2 are as follows: Step S21, method for determining the time window: The earliest hail time in the research area is advanced 2 hours as the start time, and the latest hail time is extended 2 hours as the end time. The continuous period during this time is determined as the time window of the radar data; Step S22, time downscaling processing of radar data: Assume that the evolution of radar echoes at two time intervals of 6 minutes follows a linear change law in a short time. Linear interpolation is performed on the radar data at adjacent time intervals to obtain three-dimensional radar reflectivity data with a time resolution of 1 minute, realizing the time downscaling processing of radar data. The present invention is further configured that the specific sub-steps of step S3 are as follows: Step S31, vectorization of the hail influence range: Convert the radar polar coordinate data into an equal longitude and latitude coordinate system, and interpolate the data onto a fixed equal longitude and latitude grid coordinate system; for each hail point, define a circular area with a radius of 1.5 km centered on this point; in the equal longitude and latitude grid coordinate system, mark all grid points within this circular area as the hail influence range. The marking method can adopt the method of assigning 1 / 0 to obtain the vectorized data of the hail influence range; Step S32, method for determining the threshold of hail identification index: Construct an identification index based on the height difference between the echo top height of the radar strong center (reflectivity factor ≥ 45 dBZ) and the 0°C layer height on the same day. The method for determining the threshold of this "height difference" is specifically as follows: Calculate the height differences where the top height of the strong echo (reflectivity factor ≥ 45 dBZ) is greater than the 0°C layer height on the same day, which are >0.2 km, >0.4 km... >6 km, and obtain the proportion of the cumulative hit quantity for different thresholds. Plot the hail occurrence probability corresponding to different height differences where the echo top height of 45 dBZ in the local hail cases is greater than the 0°C layer height, that is, the proportion of the hail assessment result under the current threshold setting condition in the total hit quantity. And for each threshold, the ratio of the total area of the hail influence grid inverted to the area of the hail point is calculated. Select the optimal height difference threshold with a relatively high proportion of the hit quantity under the condition of a relatively low ratio as the basis for the qualitative hail identification index.

[0010] The present invention is further configured that the hail identification index is as follows: The echo top height of the strong radar echo center (reflectivity factor ≥ 45 dBZ) reaches a certain height above the 0°C layer height of the day, and this height threshold is determined according to step S32.

[0011] The present invention is further configured as: The specific sub-steps of the said step S5 are: Step S51, Hail diameter algorithm: Judging according to the hail identification index, calculating the maximum expected hail size for the area reaching the threshold, and the calculation method refers to the formula for calculating the maximum expected hail size in the hail detection algorithm:

[0012] Where: , is the weight function of the atmospheric temperature stratification, when When The value is , is the 0°C layer height, is the reflectivity factor weight function, used to define the conversion area of the reflectivity factors of rain and hail. Judging according to step S3, when reaching the identification index threshold = 1, otherwise = 0; Step S52, Cumulative hail time: Calculating the three-dimensional radar reflectivity factor minute by minute within the time window according to the hail identification index, and counting the number of times reaching the index. Among the times reaching the standard, according to the hail diameter segments corresponding to the hail in the hail grade division, count the cumulative times of hailfall of different intensity levels. The cumulative times of hailfall of each intensity level are the corresponding cumulative hail times of different intensity hails, and the unit is minutes; Step S53, Hail process intensity algorithm: Statistically calculate the expected maximum hail diameter and the cumulative hail time T at each grid point within the time window, and determine the hail process intensity index S. The specific algorithm is:

[0013] In the formula represents the expected maximum hail diameter (unit: millimeter), and T represents the cumulative hail time (unit: minute).

[0014] The present invention is further configured as: The disaster loss rate in the said step 6 ( The determination method is as follows: Select the disaster loss data of at least three historical periods in the same period or monthly in the research area, count the cumulative amount of hail disaster losses in the research area and its ratio to the total amount. According to the historical dataset of refined hail process intensity indices obtained in step S5, count the cumulative intensity index of the hail process during the disaster loss period and conduct regional analysis to obtain the cumulative amount of hail intensity indices corresponding to the statistical period and scope of the disaster loss data in each region. Compared with the prior art, the beneficial effects of this solution are as follows: Starting from aspects such as radar mosaics, recognition indicators, and the cumulative effect of time, the present invention improves the post-disaster hail assessment method, and the obtained results of hail intensity assessment are more accurate, which can provide a scientific decision-making basis for the government's disaster prevention and mitigation work, and also has great potential in developing industry service platforms in conjunction with relevant demand departments such as the agriculture and insurance sectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is the flowchart of the intensity assessment method in an embodiment of the present invention; Figure 2 is the distribution map of hail hitting points observed on March 30, 2024 in an embodiment of the present invention; Figure 3 is the distribution map of the maximum hail diameter of hail hitting points observed on March 30, 2024 in an embodiment of the present invention; Figure 4 is the evaluation result of hail process parameters on March 30, 2024 in an embodiment of the present invention, where in Figure (a) is the maximum hail diameter, (b) is the cumulative duration, (c) is the cumulative equivalent time, and (d) is the process intensity level; Figure 5 is the comparison of the distribution of hail hitting points observed on March 30, 2024 and the evaluation result of the process intensity level in an embodiment of the present invention (the areas with letter labels correspond to Figure 1 ); Figure 6 is the distribution map of the regional hail intensity index, direct economic loss, and the proportion of disaster losses in Guiyang City from 2021 to 2023 in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] In order to enable those skilled in the art of the present technology to better understand the solution of the present invention, the technical solution of the present invention will be further described in detail below in conjunction with the embodiments and drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0017] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the embodiments.

[0018] Embodiment: A refined hail risk assessment method based on Doppler weather radar data. As Figures 1-5 shown, collect radar data, sounding data, hailfall data, and hail disaster loss data within the province. According to the national standard "Hail Grade (GBT27957-2011)", divide the hail into three grades according to the hail diameter. Classify large hailstones and extra-large hailstones into the severe grade, small hailstones corresponding to the mild grade of hail, and medium hailstones corresponding to the moderate grade of hail.

[0019] According to the time recorded by the hailfall time within the research area, determine the time window for calculating the radar data. Assume that the radar echoes at two adjacent times follow a linear change law in a short period of time, and perform linear interpolation on the radar data at two adjacent times to obtain the radar three-dimensional reflectivity factor data per minute.

[0020] Using historical hailfall event records and radar data, construct an identification index based on the height difference between the echo top height of the radar strong center (reflectivity factor ≥ 45 dBZ) and the 0°C layer height on the same day. The method for determining the threshold of this "height difference" is specifically as follows: Calculate the height differences where the top height of the strong echo (reflectivity factor ≥ 45 dBZ) is greater than the 0°C layer height on the same day, which are >0.2 km, >0.4 km... >6 km. Obtain the proportion of the cumulative hit numbers for different thresholds, and draw the hailfall probability corresponding to different height differences where the echo top height of 45 dBZ obtained from the local hail case statistics is greater than the 0°C layer height, that is, the proportion of the hail assessment results under the current threshold setting condition in the total hit numbers, and the ratio of the total area of the hail impact grid inversed by each threshold to the area of the hailfall point. Select the optimal height difference threshold with a higher proportion of hit numbers under the condition of a lower ratio as the basis for the qualitative hail identification index.

[0021] Calculate the disaster loss rates corresponding to different disaster intensity levels: According to the evaluation object, select the evaluation factors for hail disaster losses, and combine the total value of the evaluation factors in the year when the disaster occurred to calculate the disaster loss rates of each historical disaster process. According to the probability density distribution function of the disaster loss rate, use the standard deviation method to divide the disaster loss rates corresponding to different hail disaster intensities.

[0022] First, judge according to the hail identification index, and calculate the maximum expected hail size for the area reaching the threshold. The calculation method refers to the calculation formula of the maximum expected hail size (MESH) in the Hail Detection Algorithm (HDA).

[0023] Where: is the weight function of the atmospheric temperature stratification. When is the value is , is the height of the 0°C layer, is the reflectivity factor weight function, which is used to define the conversion zone of the rain and hail reflectivity factors. According to the judgment in step S3, when the recognition index threshold is reached = 1, otherwise = 0.

[0024] According to the hail recognition index, the three-dimensional radar reflectivity factor per minute within the time window is calculated, and the number of times reaching the index is counted. Among the times reaching the standard, according to "Hail Grade (GBT 27957-2011)", the cumulative times of hailfall at different intensity levels are counted.

[0025] Count the expected maximum hail diameter and the cumulative hailfall time T at each grid point within the time window, and determine the hail process intensity index S. The specific algorithm is as follows:

[0026] In the formula represents the expected maximum hail diameter (unit: millimeter), and T represents the cumulative hailfall time (unit: minute).

[0027] The hail disaster risk index H is calculated according to the following formula:

[0028] Among them, H is the process intensity index evaluated by radar data, n represents all the times within the hail process time window, and the frequencies of each disaster-causing factor reaching the thresholds of mild, moderate, and severe disaster-causing factors are respectively , and their disaster loss rates of mild, moderate, and severe disasters are respectively .

[0029] The determination method of the disaster loss rate ( ) is as follows: Select the disaster loss data of at least three historical same-period arbitrary time periods or monthly in the study area for statistics, count the cumulative amount of hail disaster losses in the study area and its ratio to the total amount, and according to the historical data set of the refined intensity index of the hail process obtained in step S5, count the cumulative intensity index of the hail process during the disaster loss period and conduct regional analysis to obtain the cumulative amount of the hail intensity index corresponding to the statistical time period and range of the disaster loss data in each region.

[0030] The total amount or value amount per unit area of the evaluation object exposed to the hail disaster risk is used as the exposure assessment index (E) after normalization. Such as the total GDP per unit area and the distribution of crop areas.

[0031] Construct an evaluation model for the refined hail disaster risk index R;

[0032] In the formula, R is the hail disaster risk index, h and e are weight coefficients respectively, h + e = 1, which can be determined by the expert scoring method. According to the standard deviation method, the hail disaster risk index is divided into five levels, corresponding to high, relatively high, medium, relatively low, and low risk levels respectively.

[0033] Select the cumulative data of hail disaster losses in Guiyang City from 2021 to 2023 for three consecutive years (as shown in Table 1). According to the distribution of the annual regional average hail index, economic losses, and the proportion of disaster losses, select a linear equation for fitting (such as Figure 6 ), and the regional disaster loss rate of annual hail disasters in Guiyang can be obtained as 0.0011, that is, 0.11%.

[0034] Table 1 Comparison table of regional hail intensity index, direct economic loss, and agricultural loss in Guiyang City from 2021 to 2023 Year Hail Index Agricultural Loss (10,000 yuan) Economic Loss (100 million yuan) Annual GDP (100 million yuan) Proportion of Disaster Loss (‰) 2021 394 14374.38 1.886 4674.761 0.404 2022 214 6735.43 0.673 4921.172 0.137 2023 493 16901.76 2.219 5154.75 0.431 From 15:00 to 21:00 on March 30, 2024, hail was observed at 135 points across the province, mainly distributed in four regions in the southern and central - southern parts of the province (such as Figure 1 ). The area with the strongest hail is Area A, which contains two strong hail belts that are very close and have the same movement direction. The maximum hail diameter recorded reached 69 mm. Judging from the combined hail diameter and duration, there were 54 points of mild hail, 32 points of moderate hail, 40 points of heavy hail, and 9 points of extremely heavy hail; the distribution map of the maximum hail diameter (such as Figure 2 ). According to the refined evaluation method of the hail process, the maximum hail diameter, cumulative duration, and cumulative equivalent time are calculated (such as Figure 3 ). In this process, the calculated maximum hail diameter reached 50 mm, the cumulative duration was up to 7 min at most, and the cumulative equivalent time reached 50 min. The process intensity evaluation level distribution is judged according to the cumulative equivalent time, and the cumulative equivalent time is obtained by weighted summing of the two - factor distribution characteristics of the combined maximum hail diameter and duration. The distribution of manually observed hail - falling points is analyzed in contrast with the refined intensity evaluation level distribution map of the hail process (such as Figure 4 ). For the four hail - falling concentrated areas, the hail - falling locations observed correspond to evaluation areas with hail of mild degree or above. The areas where the observation points are heavy and extremely heavy hail points also correspond to evaluation areas of moderate to heavy levels. Therefore, the evaluation results are reliable in both qualitative judgment (whether a hail event occurs) and quantitative intensity grading of the hail process.

[0035] The above specific embodiments are merely explanations of the present invention, and they are not limitations on the present invention. After reading this specification, those skilled in the art can make modifications to these embodiments without creative contributions as needed, but as long as they are within the scope of the claims of the present invention, they are protected by the patent law.

Claims

1. A refined risk assessment method for hail based on Doppler weather radar data, characterized in that: The following steps are involved: Step S1, data collection and preprocessing: collecting radar data, sounding data, hail data and hail disaster loss data, and dividing the hail data into three intensity levels according to the hail level classification method; Step S2, radar data preprocessing: determining the time window of radar data used for calculation according to the time records of historical hail events, interpolating the data of two adjacent time periods to obtain data with a time resolution of 1 minute; Step S3, establishing a refined hail process intensity historical data set: according to the hail identification index, the expected maximum hail diameter algorithm and the time integration method, the duration and maximum hail diameter of the hail process in the study area are calculated using the minute-by-minute radar three-dimensional reflectivity factor data, and the refined hail process intensity distribution data set in the study area is obtained according to the hail grade classification method. The specific sub-steps of step S3 are as follows: Step S31, gridded marking data of hail impact range: using the longitude and latitude grid coordinate system of radar data, marking all grid points within the circular area as hail impact range, and the marking method may adopt the method of assigning 1 / 0 to obtain gridded marking data of hail impact range; Step S32, method for determining threshold value of hail identification index: construct identification index according to the difference between echo top height of radar strong center (reflectivity factor ≥ 45dBZ) and 0℃ layer height on the day, and method for determining the "height difference" threshold is specifically as follows: calculate the height difference of strong echo (reflectivity factor ≥ 45dBZ) top height greater than 0℃ layer height on the day, which is > 0.2km, > 0.4km ... > 6km, respectively, obtain the proportion of cumulative hit number of different thresholds, draw the hail probability corresponding to different height differences of echo top height of 45dBZ greater than 0℃ layer height obtained by statistics of local hail cases, that is, the proportion of hail assessment results to the total hit number under the current threshold setting condition, and the ratio of the total area of ​​hail-affected grid to the area of ​​hail point obtained by inversion of each threshold, select the optimal height difference threshold with a higher proportion of hit number when the ratio is lower as the basis of qualitative hail identification index; Step S4, calculating the disaster loss rate corresponding to different disaster intensity levels: selecting the assessment factor of hail disaster loss according to the assessment object, combining the total value of the assessment factor in the year when the disaster occurred, calculating the disaster loss rate of each historical disaster process, and using the standard deviation method to divide the disaster loss rate corresponding to different hail disaster intensities according to the probability density distribution function of the disaster loss rate; Step S5, historical data set of refined intensity index of hail process: select the time period that reaches the hail identification index within the time window, calculate the distribution of the cumulative hailfall time and the maximum hail diameter over time within the time period, calculate the refined hail intensity index S using the time-weighted integration method, calculate the intensity index distribution of each hail process in the study area, and form a historical data set of refined intensity index of hail process; Step S6, hail disaster risk assessment: The hail disaster risk index H is calculated according to the following formula: ； Among them, H is the process intensity index evaluated by radar data, n represents all the times in the hail process time window, and the frequencies of each hazard factor reaching the thresholds of mild, moderate, and severe hazard factors are respectively The loss rates of mild, moderate and severe disasters are ; Step S7, exposure assessment of the disaster-prone object: the total amount or value per unit area of ​​the assessment object exposed to the hail disaster risk is normalized and used as the exposure assessment index (E); Step S8, hail disaster refined risk assessment and zoning: considering the hail disaster hazard and the exposure of the disaster-bearing body, an assessment model of the hail disaster refined risk index R is constructed; ； Where R is the hail disaster risk index, h and e are weight coefficients, h+e=1. It is determined according to the expert scoring method. According to the standard deviation method, the hail disaster risk index is divided into five levels, corresponding to high, relatively high, medium, relatively low and low risk levels.

2. The method for fine-grained risk assessment of hail based on Doppler weather radar data as claimed in claim 1, characterized in that: Hail is divided into small hail, medium hail, large hail and extra large hail according to its diameter: the maximum diameter of small hail is less than 5mm; the maximum diameter of medium hail is between 5~20mm, the maximum diameter of large hail is between 20~50mm, and the diameter of extra large hail is greater than 50mm. Large hail and extra large hail are classified as severe grade, small hail corresponds to light grade hail, and medium hail corresponds to moderate grade hail.

3. The method for fine-grained risk assessment of hail based on Doppler weather radar data as claimed in claim 1, characterized in that: The specific sub-steps of step S2 are: Step S21, a method for determining a time window: the earliest hailfall time in the study area is advanced by 2 hours as the start time, and the latest hailfall time is extended by 2 hours as the end time, and the duration in between is determined as the time window of the radar data; Step S22, time downscaling of radar data: Assuming that the evolution of radar echoes at two times with an interval of 6 minutes follows a linear change law in a short period of time, the radar data at adjacent times are linearly interpolated to obtain three-dimensional radar reflectivity data with a time resolution of 1 minute, thereby realizing time downscaling of radar data.

4. The method for fine-grained risk assessment of hail based on Doppler weather radar data as claimed in claim 3, characterized in that: The hail identification indicators are as follows: The echo top height of the radar strong echo center (reflectivity factor ≥ 45dBZ) reaches a certain height above the 0°C layer height on that day, and the height threshold is determined according to step S32.

5. The method for fine-grained risk assessment of hail based on Doppler weather radar data as claimed in claim 2, characterized in that: The specific sub-steps of step S5 are: Step S51, hail diameter algorithm: according to the hail identification index, the maximum expected hail size is calculated for the area reaching the threshold. The calculation method refers to the calculation formula of the maximum expected hail size in the hail detection algorithm: ； in: , is the weight function of the atmospheric temperature stratification, when hour Value , is the 0℃ layer height, is the reflectivity factor weight function, which is used to define the transition zone between rain and hail reflectivity factors. According to step S3, when the identification index threshold is reached, =1, otherwise ; Step S52, cumulative hailfall time: the three-dimensional radar reflectivity factor is calculated minute by minute within the time window according to the hail recognition index, and the number of times that the index is reached is counted. Among the times that the index is reached, the cumulative times of hailfall at different intensity levels are counted in sections according to the hail diameters corresponding to the hail in the hail grade classification. The cumulative times of hailfall at each intensity correspond to the cumulative hailfall time of hailfall at different intensities, and the unit is minutes. Step S53, hail process intensity algorithm: count the expected maximum hail diameter at each grid point within the time window and the accumulated hail time T, determine the hail process intensity index S. The specific algorithm is: ； In the formula It represents the expected maximum hail diameter (unit: millimeter), and T represents the accumulated hail time (unit: minute).

6. The method for fine-grained risk assessment of hail based on Doppler weather radar data as claimed in claim 1, characterized in that: The disaster loss rate in step 6 The determination method is as follows: select statistical data of at least three arbitrary time periods or monthly disaster loss data from the same period in the history within the study area, and calculate the cumulative amount of hail disaster losses in the study area and its ratio to the total amount; based on the historical data set of the refined intensity index of the hail process obtained in step S5, calculate the cumulative intensity index of the hail process within the disaster loss period, and perform regional analysis to obtain the cumulative amount of hail intensity index corresponding to the statistical period and range of the disaster loss data in each region.

Images