A precipitation weather phenomenon identification method based on laser drop spectrum

Abstract

This invention relates to the field of meteorological observation technology, specifically to a method for identifying precipitation weather phenomena based on laser droplet spectra. It addresses the problems of existing laser droplet spectra-based precipitation instrument identification algorithms lacking unified standards, having a high misclassification rate for precipitation phases, and insufficient accuracy in identifying weak precipitation and severe convective weather. This invention simultaneously collects two-dimensional data of precipitation particle raindrop spectra and surface temperature data using a laser droplet spectra-based precipitation instrument. It delineates particle identification zones for different precipitation types using isovelocity lines and isodiameter lines, calculates core identification feature parameters such as particle number ratio, spectral width, and mode, and classifies precipitation according to priority order: hail, drizzle, precipitation, and snowfall. Phenomena not meeting the aforementioned criteria are identified as sleet. This invention complies with meteorological industry observation standards and can be directly embedded into existing compliant laser droplet spectra-based precipitation instruments, making it widely applicable to various ground meteorological precipitation observation scenarios.

Description

Technical Field

[0001] This invention relates to the field of meteorological observation technology, specifically to a method for identifying precipitation weather phenomena based on laser droplet spectrum. Background Technology

[0002] Precipitation is a core element of surface meteorological observation, and its accurate identification is of great significance for meteorological forecasting and early warning, disaster prevention and mitigation, hydrological monitoring, and ecological environment research. Laser drop spectrometers, with their advantages of high temporal resolution, high measurement accuracy, and all-weather automatic observation capabilities, have become the mainstream equipment for precipitation observation in my country's surface meteorological observation network. The performance specifications of these instruments have been defined in the meteorological industry standard QX / T565-2020, "Laser Drop Spectrometers for Precipitation."

[0003] Currently, there is no unified technical standard for precipitation identification algorithms in laser droplet precipitation instruments. Different manufacturers have significant differences in the core logic, judgment thresholds, and feature parameter selection of their algorithms, which mainly have the following technical defects:

[0004] Most algorithms rely solely on the proportion of particles as the core criterion for judgment, without combining the spectral width of precipitation particles, mode characteristics, and auxiliary factors such as ground temperature. As a result, the accuracy of identifying weak precipitation (drizzle) and mixed-phase precipitation (sleet) is low, and misjudgments of rain and drizzle or rain and snow are easy to occur.

[0005] The identification of severe convective precipitation such as hail does not set multi-dimensional constraints, and only uses particle diameter as the judgment standard, which easily misidentifies large raindrops and graupel as hail, resulting in a high false alarm rate.

[0006] There are no standardized rules for particle classification, characteristic parameter calculation, and identification and judgment procedures. The comparability of observation data from different instruments is poor, and the consistency of identification results is insufficient, which cannot meet the standardized business application requirements of the national-level ground meteorological observation network.

[0007] The existing GB / T35224-2017 "Specifications for Ground Meteorological Observation of Weather Phenomena" only specifies the manual observation specifications for precipitation weather phenomena, and does not formulate a practical quantitative algorithm standard for automatic identification by laser drop spectrometers, resulting in a large discrepancy between the results of automatic observation and manual observation.

[0008] To address the shortcomings of the existing technologies, this invention proposes a standardized, multi-factor, and highly accurate method for identifying precipitation weather phenomena, based on the raindrop spectrum detection principle of a laser drop spectrometer and combined with the observation needs of the meteorological industry, thus solving many of the pain points of the existing technologies. Summary of the Invention

[0009] (a) Technical problems to be solved

[0010] To address the shortcomings of existing technologies, this invention provides a method for identifying precipitation weather phenomena based on laser drop spectrum. Through standardized particle recognition area division, multi-dimensional feature parameter calculation, and hierarchical priority determination logic, it achieves accurate and standardized identification of five core precipitation weather phenomena: drizzle, rain, snow, sleet, and hail. This improves the consistency, accuracy, and operational availability of observation data from laser drop spectrum precipitation phenomena instruments, filling the gap in the industry for automatic precipitation phenomenon identification algorithms that lack unified standards.

[0011] (II) Technical Solution

[0012] To achieve the above objectives, the present invention provides the following technical solution: a method for identifying precipitation weather phenomena based on laser droplet spectrum, comprising collecting raindrop spectrum data of precipitation particles using a laser droplet spectrum precipitation phenomenon instrument, and classifying and identifying precipitation weather phenomena based on the raindrop spectrum data, including the following steps:

[0013] Step 1: Within the preset sampling period, two-dimensional data of raindrop spectra of precipitation particles in the sampling space are collected using a laser droplet spectrometer, and ground temperature data for the corresponding time period are collected simultaneously; the two-dimensional data of raindrop spectra includes the diameter, terminal velocity of each precipitation particle, and particle counts corresponding to different diameter and velocity categories.

[0014] Step 2: Using isotropic lines and isodiameter lines, delineate particle identification zones corresponding to different precipitation types in a two-dimensional coordinate system of particle diameter and terminal velocity; the isotropic lines include... and The equal diameter lines include 0.6 mm and 5 mm; the particle identification area includes drizzle particle identification area, rain particle identification area, snow particle identification area, mixed precipitation particle identification area, graupel particle identification area and hail particle identification area, wherein the graupel particle identification area is only used as a reference area for hail weather determination;

[0015] Step 3: Based on the particle identification area defined in Step 2 and the two-dimensional raindrop spectrum data collected in Step 1, calculate the precipitation identification feature parameters within the sampling period; the feature parameters include the total number of particles within the sampling period, the number of particles corresponding to each particle identification area and the proportion of the number of particles, the particle diameter spectrum width, the particle velocity spectrum width, the dominant particle diameter, and the dominant particle velocity.

[0016] Step 4: First, use the ground temperature data collected in Step 1 to perform a preliminary screening of precipitation phases. Then, according to the priority order of hail, drizzle, rain, and snow, perform precipitation weather phenomenon determination in sequence based on the feature parameters calculated in Step 3 and the ground temperature data. If the precipitation characteristics in this sampling period do not meet the determination conditions of any of the aforementioned precipitation types, it is identified as sleet.

[0017] Furthermore, in step 4, the criteria for determining hail weather are: within the sampling period, the number of precipitation particles with a diameter ≥ 5 mm is ≥ 6, and simultaneously the ratio of hail particles to graupel particles is ≥ 0.3, the particle diameter spectrum is ≥ 5 mm, and the particle velocity spectrum is ≥ 0.3. The particle diameter corresponding to the velocity spectrum width is ≥5mm.

[0018] Furthermore, in step 4, the criterion for determining drizzle weather is that any of the following conditions are met:

[0019] Condition 1: During the sampling period, the proportion of precipitation particles with a diameter <0.6mm is ≥50%;

[0020] Condition 2: During the sampling period, the total number of particles from rain and drizzle accounts for ≥50%, or the total number of particles from snow and drizzle accounts for ≥50%, and simultaneously meets the requirements of particle diameter spectral width <3.0mm and particle velocity spectral width. .

[0021] Furthermore, in step 4, the criteria for determining rainfall weather are: ground temperature > 2℃, and meeting any of the following conditions:

[0022] Condition 1: During the sampling period, the proportion of particles in the rain particle recognition area is ≥50%;

[0023] Condition 2: During the sampling period, the proportion of particles in the rain particle recognition area is ≥30%, and the mode velocity of particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥1.2mm, or the particle velocity corresponding to the diameter spectrum width is... .

[0024] Furthermore, in step 4, the criteria for determining snowfall weather are: ground temperature ≤ 2℃, and meeting any of the following conditions:

[0025] Condition 1: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥50%;

[0026] Condition 2: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥30%, and the mode velocity of the particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥3mm, or the particle velocity corresponding to the diameter spectrum width is ≥3mm. .

[0027] Further, in step 3, the particle count percentage is the percentage of the number of particles in the corresponding particle recognition area to the total number of particles in the sampling period; the particle diameter spectrum width is the maximum diameter of precipitation particles in the sampling period; the particle velocity spectrum width is the maximum falling velocity of precipitation particles in the sampling period; the particle mode diameter is the diameter value corresponding to the diameter tier with the most particle counts in the sampling period; and the particle mode velocity is the falling velocity value corresponding to the velocity tier with the most particle counts in the sampling period.

[0028] Further, in step 1, the particle diameter and terminal velocity of the raindrop spectrum two-dimensional data are quantized using 32 channels; wherein, the particle diameter range is from 0.062 mm to 24.5 mm, and the interval between quantizations increases gradually from 0.125 mm to 3.0 mm as the diameter increases; the terminal velocity range is... to The interval between gears increases with speed from Gradient increment to .

[0029] Furthermore, the specific rules for the preliminary screening of precipitation phases in step 4 are as follows: when the ground temperature is >2℃, the determination of liquid precipitation weather phenomena is given priority; when the ground temperature is ≤2℃, the determination of solid and mixed precipitation weather phenomena is given priority.

[0030] Furthermore, in step 2, the defined graupel identification area is only used as a reference area for hail weather determination, and graupel is not output as a separate identification result of precipitation weather phenomenon.

[0031] (III) Beneficial Effects

[0032] Compared with existing technologies, this invention provides a method for identifying precipitation weather phenomena based on laser droplet spectra, which has the following beneficial effects:

[0033] This invention, based on laser drop spectrum, is a precipitation weather phenomenon identification method. It employs a composite logic of particle partitioning, multiple feature parameters, temperature auxiliary factors, and hierarchical priority determination, breaking through the technical bottleneck of existing single-factor determination. It effectively solves the problem of misjudging drizzle and rain, rain and snow, and hail and sleet. According to actual operational tests, the consistency with human observation results is over 95%. In particular, the accuracy of identification of mixed-phase precipitation, weak precipitation, and strong convective hail is improved by more than 10% compared with existing algorithms.

[0034] The particle classification, feature parameter definition, judgment threshold, and identification process of this invention strictly follow the relevant national standards and meteorological industry standards for ground meteorological observation. It can be directly embedded into existing laser drop spectrum precipitation phenomena instruments that conform to the QX / T565-2020 standard, realizing unified processing and result comparison of observation data from instruments of different manufacturers, filling the gap in the industry for automatic precipitation phenomenon identification algorithms without unified standards.

[0035] This invention establishes a hierarchical priority judgment mechanism, prioritizing the identification of hail weather with significant characteristics, followed by weak precipitation such as drizzle that is easily obscured, then distinguishing rain and snow phases through temperature factor constraints, and finally including sleet as a fallback, avoiding missed and false alarms. Simultaneously, hail judgment incorporates multi-dimensional constraints, effectively filtering out interference from large raindrops and graupel, significantly reducing the false alarm rate for hail. This invention requires no hardware modification to existing laser droplet spectral precipitation instruments; it can be implemented solely through software algorithm upgrades. It has low computational load, strong real-time performance, and is adaptable to the operational needs of national, provincial, and municipal ground meteorological observation network. It can also be applied to precipitation monitoring scenarios in hydrology, transportation, civil aviation, and other industries. All feature parameter calculations and judgment processes in this invention have clear quantification rules, and the identification results can be verified through backtesting of original raindrop spectral data, meeting the quality control and operational auditing requirements of meteorological observation data. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the raindrop spectrum distribution for precipitation particle identification according to the present invention;

[0037] Figure 2 This is a raindrop spectrum distribution diagram of a rainy weather event in Embodiment 1 of the present invention;

[0038] Figure 3 This is a raindrop spectrum distribution diagram of snowfall weather in Embodiment 2 of the present invention;

[0039] Figure 4 This is a raindrop spectrum distribution diagram of hail weather in Embodiment 3 of the present invention;

[0040] Figure 5 This is a raindrop spectrum distribution diagram of sleet weather in Embodiment 4 of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Please see Figures 1 to 5This invention relates to a method for identifying precipitation weather phenomena based on laser droplet spectrum, comprising collecting raindrop spectrum data of precipitation particles using a laser droplet spectrum precipitation instrument, and classifying and identifying precipitation weather phenomena based on the raindrop spectrum data, including the following steps:

[0043] Step 1: Within the preset sampling period, two-dimensional data of raindrop spectra of precipitation particles in the sampling space are collected using a laser droplet spectrometer, and ground temperature data for the corresponding time period are collected simultaneously; the two-dimensional data of raindrop spectra includes the diameter, terminal velocity of each precipitation particle, and particle counts corresponding to different diameter and velocity categories.

[0044] Step 2: Using isotropic lines and isodiameter lines, delineate particle identification zones corresponding to different precipitation types in a two-dimensional coordinate system of particle diameter and terminal velocity; the isotropic lines include... and The equal diameter lines include 0.6 mm and 5 mm; the particle identification area includes drizzle particle identification area, rain particle identification area, snow particle identification area, mixed precipitation particle identification area, graupel particle identification area and hail particle identification area, wherein the graupel particle identification area is only used as a reference area for hail weather determination;

[0045] Step 3: Based on the particle identification area defined in Step 2 and the two-dimensional raindrop spectrum data collected in Step 1, calculate the precipitation identification feature parameters within the sampling period; the feature parameters include the total number of particles within the sampling period, the number of particles corresponding to each particle identification area and the proportion of the number of particles, the particle diameter spectrum width, the particle velocity spectrum width, the dominant particle diameter, and the dominant particle velocity.

[0046] Step 4: First, use the ground temperature data collected in Step 1 to perform a preliminary screening of precipitation phases. Then, according to the priority order of hail, drizzle, rain, and snow, perform precipitation weather phenomenon determination in sequence based on the feature parameters calculated in Step 3 and the ground temperature data. If the precipitation characteristics in this sampling period do not meet the determination conditions of any of the aforementioned precipitation types, it is identified as sleet.

[0047] In this scheme, the criteria for determining hail weather in step 4 are: within the sampling period, the number of precipitation particles with a diameter ≥ 5 mm is ≥ 6, and the ratio of hail particles to graupel particles simultaneously satisfies ≥ 0.3, particle diameter spectral width ≥ 5 mm, and particle velocity spectral width... The particle diameter corresponding to the velocity spectrum width is ≥5mm. This completely solves the industry pain point that existing technologies only judge hail based on particle diameter, which easily misclassifies large raindrops and graupel as hail, significantly reducing the false hail alarm rate. Actual measurements show that the hail recognition accuracy is more than 20% higher than traditional algorithms. The multi-dimensional constraints form a rigorous judgment logic, which not only avoids noise interference from sporadic large particles, but also accurately captures the core characteristics of severe convective hail weather, without missing weak hail processes. At the same time, it is consistent with the national standard GB / T27957-2011 "Hail Classification" and complies with meteorological business specifications. As the highest priority judgment item, it prioritizes the most significant severe convective weather characteristics, avoids subsequent low-priority judgments from covering hail characteristics, and eliminates the possibility of missing hail.

[0048] In this scheme, in step 4, the criterion for determining drizzle weather is that any of the following conditions are met:

[0049] Condition 1: During the sampling period, the proportion of precipitation particles with a diameter <0.6mm is ≥50%;

[0050] Condition 2: During the sampling period, the total number of particles from rain and drizzle accounts for ≥50%, or the total number of particles from snow and drizzle accounts for ≥50%, and simultaneously meets the requirements of particle diameter spectral width <3.0mm and particle velocity spectral width. The core physical characteristics of drizzle are extremely small particle diameter (generally <0.6mm), uniform particle size, slow falling speed, and narrow spectral width, which distinguishes it from rainy weather with larger droplet diameter and wider spectral width. The main judgment condition (condition one) directly locks in the core small diameter characteristic of drizzle, using the proportion of <0.6mm particles ≥50% as the core judgment criterion, which is consistent with the microphysical nature of drizzle. The supplementary judgment condition (condition two) is designed for weak mixing scenarios of drizzle and rain / snow. It constrains the uniformity of particles through a spectral width threshold, avoiding misjudging weak rainfall or weak snowfall as drizzle, and at the same time solving the problem of missed judgment in weak mixing scenarios under the pure drizzle judgment condition.

[0051] In this scheme, the criteria for determining rainfall weather in step 4 are: ground temperature > 2℃, and meeting any of the following conditions:

[0052] Condition 1: During the sampling period, the proportion of particles in the rain particle recognition area is ≥50%;

[0053] Condition 2: During the sampling period, the proportion of particles in the rain particle recognition area is ≥30%, and the mode velocity of particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥1.2mm, or the particle velocity corresponding to the diameter spectrum width is... The core physical characteristic of liquid precipitation is that particle diameter is positively correlated with falling velocity; the larger the raindrop diameter, the higher the terminal velocity. Simultaneously, the formation and maintenance of liquid precipitation require near-surface temperatures above 0°C. The industry-standard critical temperature for liquid precipitation is 2°C; below this temperature, supercooled water and solid precipitation are more likely to occur. By first imposing a temperature constraint, the environmental conditions for liquid precipitation formation are locked in, preventing the misclassification of snowfall or sleet as precipitation in low-temperature environments. The primary judgment condition directly identifies pure precipitation scenarios based on the proportion of rain particles. Supplementary judgment conditions address weakly mixed scenarios of rain with other phases, using multi-dimensional parameters such as mode velocity and spectral width to reconstruct the core characteristic of a positive correlation between precipitation velocity and diameter, avoiding missed detections in weak precipitation scenarios.

[0054] In this scheme, the criteria for determining snowfall weather in step 4 are: ground temperature ≤ 2℃, and meeting any of the following conditions:

[0055] Condition 1: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥50%;

[0056] Condition 2: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥30%, and the mode velocity of the particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥3mm, or the particle velocity corresponding to the diameter spectrum width is ≥3mm. The core physical characteristics of solid snowfall are that snowflakes are mostly flake-shaped and have low density. Even if they are large in diameter, their terminal velocity is much lower than that of raindrops of the same diameter, generally ≤3 m·s⁻¹. At the same time, the formation and maintenance of snowfall require the near-surface temperature to be below a critical value. The industry-standard critical temperature for solid precipitation is 2℃. First, by using temperature as a pre-constraint, the environmental conditions for the formation of solid precipitation are locked in, avoiding the misjudgment of large raindrops as snowflakes in high-temperature environments. The main judgment condition directly locks in pure snowfall scenarios by the proportion of snow particles. Supplementary judgment conditions are used for weakly mixed scenarios of snow and other phases. Through core characteristics such as low mode velocity and narrow velocity spectrum, the microphysical nature of low density and low falling velocity of snowflakes is restored, avoiding missed judgments in weak snowfall scenarios. The pre-constraint of temperature fundamentally solves the core pain point of misjudging the phase of rain and snow, improving the accuracy of snowfall phase recognition by more than 15% compared with traditional algorithms; the dual-condition judgment logic takes into account both heavy and light snowfall scenarios, adapts to the full range of snowfall recognition from light snowfall to blizzards, and conforms to the operational specifications of ground meteorological observation; the multi-parameter constraint conditions accurately distinguish between large-diameter snowflakes and large raindrops, solving the problem that traditional algorithms easily misjudge large-diameter snowflakes as raindrops, and significantly reducing the misjudgment rate of snowfall.

[0057] In this scheme, in step 3, the particle count percentage is the percentage of the number of particles in the corresponding particle recognition area to the total number of particles in the sampling period; the particle diameter spectrum width is the maximum diameter of precipitation particles in the sampling period; the particle velocity spectrum width is the maximum terminal velocity of precipitation particles in the sampling period; the particle mode diameter is the diameter value corresponding to the diameter tier with the most particle counts in the sampling period; and the particle mode velocity is the terminal velocity value corresponding to the velocity tier with the most particle counts in the sampling period. The core of the laser drop spectrometer for precipitation phenomena is to achieve quantitative measurement of particle diameter and velocity through channel grading. The accuracy and range of grading directly determine the accuracy of precipitation identification. In existing technologies, the channel grading of different manufacturers varies greatly, resulting in data incompatibility. This invention unifies the core measurement channel grading rules of the laser drop spectrometer, realizing the interoperability and universality of raw data from different instruments, filling the gap in the industry for standardized grading specifications. The gradient grading design takes into account both the identification accuracy of weak precipitation and the measurement range of strong convective precipitation, ensuring the identification accuracy of small particles such as drizzle, while fully covering the measurement needs of large-diameter, high-velocity particles such as hail.

[0058] In this scheme, in step 1, the particle diameter and terminal velocity of the raindrop spectrum two-dimensional data are quantized using 32 channels; wherein, the particle diameter range is from 0.062mm to 24.5mm, and the interval between quantizations increases gradually from 0.125mm to 3.0mm as the diameter increases; the terminal velocity range is... to The interval between gears increases with speed from Gradient increment to The formation of precipitation phases is directly related to near-surface temperature. Temperature is the core environmental factor distinguishing between liquid and solid precipitation. This invention first uses temperature for initial phase screening, which can significantly reduce the judgment process and avoid misjudgment of phases from an environmental logic perspective. When the temperature is >2℃, liquid precipitation is a high-probability event, and liquid precipitation judgment is performed first, which can quickly identify rain and drizzle scenarios. When the temperature is ≤2℃, solid / mixed precipitation is a high-probability event, and the corresponding judgment is performed first, which can quickly identify snow and sleet scenarios.

[0059] In this solution, the specific rules for the initial screening of precipitation phases in step 4 are as follows: when the ground temperature is >2℃, the determination of liquid precipitation weather phenomena is prioritized; when the ground temperature is ≤2℃, the determination of solid and mixed precipitation weather phenomena is prioritized. This significantly improves the algorithm's operating efficiency, reduces unnecessary judgment processes, and adapts to the real-time operational needs of meteorological observation network. It further reduces the misjudgment rate of rain and snow phases from the environmental factor level, forming a dual judgment logic of "initial environmental screening + fine judgment based on particle features," further improving the recognition accuracy. It conforms to the basic laws of precipitation phase formation in atmospheric physics, the algorithm logic is more rigorous, consistent with the experience logic of manual observation, and improves the acceptance of automatic identification results by operational personnel.

[0060] In this scheme, the graupel identification area in step 2 is only used as a reference area for hail weather determination, and graupel is not output as a separate precipitation weather phenomenon. This avoids confusion between graupel and snowflakes or hail, significantly reduces the misclassification rate of precipitation types, and simplifies the business application process. Retaining the graupel area as a reference for hail determination does not affect the accuracy of hail identification, and avoids invalid identification item settings, thus balancing the rigor of the algorithm with business practicality.

[0061] Particle diameter and velocity 32-channel quantization table:

[0062]

[0063]

[0064] Example 1: Identification of Rainy Weather

[0065] The test data in this embodiment comes from 1-minute sampling data from the Wuchuan County Meteorological Observatory in Guizhou Province, from 01:23 to 01:24 on April 13, 2018, with a synchronous ground temperature of 4.2℃. The specific identification process is as follows:

[0066] Data Acquisition: Two-dimensional raindrop spectrum data for this period were collected using a laser droplet spectrometer, with a total of 1862 particles. The ground temperature was simultaneously collected at 4.2℃ > 2℃, and liquid precipitation was prioritized for determination.

[0067] Particle identification zone matching: The number of particles falling into each identification zone was counted. The number of rain particles in the identification zone was 1154, the number of drizzle particles in the identification zone was 628, and there were no snow, sleet, or hail particles.

[0068] Characteristic parameter calculation:

[0069] Rain particle number percentage = 1154 / 1862 × 100% ≈ 61.98%;

[0070] Particle mode velocity Particle diameter spectral width = 3.75 mm, particle velocity spectral width ;

[0071] The diameter corresponding to the velocity spectral width is 2.125 mm, and the velocity corresponding to the diameter spectral width is... .

[0072] Grading determination:

[0073] Highest priority hail judgment: No particles with a diameter ≥ 5mm, the judgment condition is not met;

[0074] Second-priority drizzle determination: The percentage of particles with a diameter < 0.6 mm = 628 / 1862 × 100% ≈ 33.73% < 50%, which does not meet condition one; The combined percentage of rain + drizzle is 100% ≥ 50%, but the diameter spectrum width is 3.75 mm ≥ 3.0 mm, which does not meet condition two, so drizzle is excluded.

[0075] The third priority for rainfall determination is: ground temperature 4.2℃ > 2℃, and the percentage of rain particles ≥ 50% (61.98%), which meets the first condition for rainfall determination.

[0076] Identification results: The precipitation weather phenomenon in this sampling period is rainfall, which is consistent with the results of manual observation during the same period.

[0077] Example 2: Identification of Snowfall Weather:

[0078] The test data in this embodiment comes from a 1-minute sampling data collected by the Weining County Meteorological Observatory in Guizhou Province from 21:44 to 21:45 on January 24, 2020, with a synchronous ground temperature of -0.8℃. The specific identification process is as follows:

[0079] Data Acquisition: Two-dimensional raindrop spectrum data for this period were collected, with a total of 986 particles. The ground temperature was -0.8℃ to 2℃. Solid precipitation was prioritized for determination.

[0080] Particle identification zone matching: The number of particles falling into each identification zone was counted. Among them, the number of snow particles in the snow particle identification zone was 592, the number of drizzle particles in the drizzle particle identification zone was 394, and there were no rain, sleet, or hail particles.

[0081] Characteristic parameter calculation:

[0082] The percentage of snow particles = 592 / 986 × 100% ≈ 60.04%;

[0083] Particle mode velocity Particle diameter spectral width = 4.25 mm, particle velocity spectral width ;

[0084] The diameter corresponding to the velocity spectral width is 3.25 mm, and the velocity corresponding to the diameter spectral width is... .

[0085] Grading determination:

[0086] Highest priority hail judgment: No particles with a diameter ≥ 5mm, the judgment condition is not met;

[0087] Second-priority drizzle determination: The percentage of particles with a diameter < 0.6 mm = 394 / 986 × 100% ≈ 39.96% < 50%, which does not meet condition one; the combined percentage of snow + drizzle is 100% ≥ 50%, but the diameter spectrum width is 4.25 mm ≥ 3.0 mm, which does not meet condition two, so drizzle is excluded.

[0088] The third priority for precipitation determination: if the ground temperature is -0.8℃ to 2℃, the temperature prerequisite for precipitation is not met, and precipitation is excluded.

[0089] Fourth priority snowfall determination: Ground temperature -0.8℃≤2℃, snow particle count ≥50% (60.04%), meeting snowfall determination condition one.

[0090] Identification results: The precipitation weather phenomenon in this sampling period is snowfall, which is consistent with the results of manual observation during the same period.

[0091] Example 3: Identification of Hail Weather:

[0092] The test data in this embodiment comes from a 1-minute sampling data collected by the Weining County Meteorological Observatory in Guizhou Province from 21:35 to 21:36 on April 14, 2023. The synchronous ground temperature was 8.6℃. The specific identification process is as follows:

[0093] Data acquisition: Two-dimensional raindrop spectrum data were collected during this period, with a total of 2458 particles, and the ground temperature was 8.6℃.

[0094] Particle identification zone matching: The number of particles falling into each identification zone is counted. The number of hail particles in the identification zone is 12, the number of graupel particles in the identification zone is 22, the number of rain particles in the identification zone is 2186, and the number of drizzle particles in the identification zone is 238.

[0095] Characteristic parameter calculation:

[0096] Number of particles with a diameter ≥ 5mm = 34 particles ≥ 6 particles;

[0097] Number of hail particles / number of graupel particles=12 / 22≈0.55≥0.3;

[0098] Particle diameter spectral width = 8.5mm ≥ 5mm, particle velocity spectral width ;

[0099] The diameter corresponding to the velocity spectrum width is 8.5mm ≥ 5mm.

[0100] Grading determination:

[0101] Highest priority hail determination: If all the constraints for hail determination are met, it is directly identified as hail weather, and no further low priority determination is required.

[0102] Identification results: The precipitation weather phenomenon in this sampling period was hail, which is consistent with the results of manual observation and weather radar monitoring during the same period. There were no omissions or misjudgments.

[0103] Example 4: Identification of sleet weather:

[0104] The test data in this embodiment comes from a 1-minute sampling data collected by the Bijie Meteorological Observatory in Guizhou Province from 14:46 to 14:47 on February 6, 2022. The synchronous ground temperature was 1.2℃. The specific identification process is as follows:

[0105] Data acquisition: Two-dimensional data of raindrop spectra were collected during this period, with a total of 1256 particles, and the ground temperature was 1.2℃≤2℃.

[0106] Particle identification zone matching: The number of particles falling into each identification zone was counted. The number of particles in the rain particle identification zone was 452, the number of particles in the snow particle identification zone was 428, the number of particles in the drizzle particle identification zone was 376, and there were no sleet or hail particles.

[0107] Characteristic parameter calculation:

[0108] Rain particle count percentage = 452 / 1256 × 100% ≈ 35.99%, snow particle count percentage = 428 / 1256 × 100% ≈ 34.08%;

[0109] The percentage of particles with a diameter < 0.6 mm = 376 / 1256 × 100% ≈ 29.94%;

[0110] Particle mode velocity Particle diameter spectral width = 3.25 mm, particle velocity spectral width .

[0111] Grading determination:

[0112] Highest priority hail judgment: No particles with a diameter ≥ 5mm, the judgment condition is not met;

[0113] Second-priority drizzle determination: The proportion of particles with a diameter < 0.6 mm is 29.94% < 50%, which does not meet condition one; the total proportion of rain + drizzle is ≈ 65.93% ≥ 50%, but the diameter spectrum width is 3.25 mm ≥ 3.0 mm, which does not meet condition two, so drizzle is excluded.

[0114] The third priority for precipitation determination: if the ground temperature is 1.2℃≤2℃, the temperature prerequisite for precipitation is not met, and precipitation is excluded.

[0115] Fourth priority snowfall determination: Snow particle percentage < 50%, not meeting condition one; Snow particle percentage ≥ 30%, but the mode velocity of particles is high. Diameter spectral width ≥ 1.5 mm, velocity spectral width The diameter corresponding to the velocity spectral width is 2.75 mm < 3 mm, and the velocity corresponding to the diameter spectral width is... Since all constraints of condition two are not met, snowfall is excluded.

[0116] Last resort: If the conditions for determining any of the aforementioned precipitation types are not met, the weather is identified as sleet.

[0117] Identification results: The precipitation weather phenomenon in this sampling period was sleet, which is consistent with the results of manual observation during the same period. This solves the problem that existing algorithms easily misclassify mixed-phase precipitation as single rain or snow.

[0118] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for identifying precipitation weather phenomena based on laser droplet spectrum, comprising collecting raindrop spectrum data of precipitation particles using a laser droplet spectrum precipitation phenomenon instrument, and classifying and identifying precipitation weather phenomena based on the raindrop spectrum data, characterized in that, Includes the following steps: Step 1: Within the preset sampling period, two-dimensional data of raindrop spectra of precipitation particles in the sampling space are collected using a laser droplet spectrometer, and ground temperature data for the corresponding time period are collected simultaneously; the two-dimensional data of raindrop spectra includes the diameter, terminal velocity of each precipitation particle, and particle counts corresponding to different diameter and velocity categories. Step 2: Using isotropic lines and isodiameter lines, delineate particle identification zones corresponding to different precipitation types in a two-dimensional coordinate system of particle diameter and terminal velocity; the isotropic lines include... and The equal diameter lines include 0.6 mm and 5 mm; the particle identification area includes drizzle particle identification area, rain particle identification area, snow particle identification area, mixed precipitation particle identification area, graupel particle identification area and hail particle identification area, wherein the graupel particle identification area is only used as a reference area for hail weather determination; Step 3: Based on the particle identification area defined in Step 2 and the two-dimensional raindrop spectrum data collected in Step 1, calculate the precipitation identification feature parameters within the sampling period; the feature parameters include the total number of particles within the sampling period, the number of particles corresponding to each particle identification area and the proportion of the number of particles, the particle diameter spectrum width, the particle velocity spectrum width, the dominant particle diameter, and the dominant particle velocity. Step 4: First, use the ground temperature data collected in Step 1 to perform a preliminary screening of precipitation phases. Then, according to the priority order of hail, drizzle, rain, and snow, perform precipitation weather phenomenon determination in sequence based on the feature parameters calculated in Step 3 and the ground temperature data. If the precipitation characteristics in this sampling period do not meet the determination conditions of any of the aforementioned precipitation types, it is identified as sleet.

2. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 4, the criteria for determining hail weather are: within the sampling period, the number of precipitation particles with a diameter ≥ 5 mm is ≥ 6, and the ratio of hail particles to graupel particles simultaneously meets the following conditions: ≥ 0.3, particle diameter spectrum width ≥ 5 mm, and particle velocity spectrum width. The particle diameter corresponding to the velocity spectrum width is ≥5mm.

3. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 4, the criterion for determining drizzle weather is that any of the following conditions are met: Condition 1: During the sampling period, the proportion of precipitation particles with a diameter <0.6mm is ≥50%; Condition 2: During the sampling period, the total number of particles from rain and drizzle accounts for ≥50%, or the total number of particles from snow and drizzle accounts for ≥50%, and simultaneously meets the requirements of particle diameter spectral width <3.0mm and particle velocity spectral width. .

4. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 4, the criteria for determining rainy weather are: ground temperature > 2℃, and meeting any of the following conditions: Condition 1: During the sampling period, the proportion of particles in the rain particle recognition area is ≥50%; Condition 2: During the sampling period, the proportion of particles in the rain particle recognition area is ≥30%, and the mode velocity of particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥1.2mm, or the particle velocity corresponding to the diameter spectrum width is... .

5. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 4, the criteria for determining snowfall weather are: ground temperature ≤ 2℃, and meeting any of the following conditions: Condition 1: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥50%; Condition 2: During the sampling period, the proportion of snow particles in the snow particle recognition area is ≥30%, and the mode velocity of the particles is also satisfied. Particle diameter spectral width ≥ 1.5 mm, particle velocity spectral width The particle diameter corresponding to the velocity spectrum width is ≥3mm, or the particle velocity corresponding to the diameter spectrum width is ≥3mm. .

6. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 3, the particle count percentage is the percentage of the number of particles in the corresponding particle recognition area to the total number of particles in the sampling period; the particle diameter spectrum width is the maximum diameter of precipitation particles in the sampling period; the particle velocity spectrum width is the maximum falling velocity of precipitation particles in the sampling period; the particle mode diameter is the diameter value corresponding to the diameter class with the most particle counts in the sampling period; and the particle mode velocity is the falling velocity value corresponding to the velocity class with the most particle counts in the sampling period.

7. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 1, the particle diameter and terminal velocity of the raindrop spectrum two-dimensional data are quantized using 32 channels; the particle diameter range is from 0.062 mm to 24.5 mm, and the interval between quantization increments from 0.125 mm to 3.0 mm as the diameter increases; the terminal velocity range is... to The interval between gears increases with speed from Gradient increment to .

8. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, The specific rules for the preliminary screening of precipitation phases in step 4 are as follows: when the ground temperature is >2℃, the determination of liquid precipitation weather phenomena is given priority; when the ground temperature is ≤2℃, the determination of solid and mixed precipitation weather phenomena is given priority.

9. The method for identifying precipitation weather phenomena based on laser droplet spectrum according to claim 1, characterized in that, In step 2, the defined graupel identification area is only used as a reference area for hail weather determination, and graupel is not output as a separate identification result of precipitation weather phenomenon.

Images