A stratiform cloud rainfall microphysical structure inversion method, artificial rainmaking potentiality discrimination method and system

Abstract

This invention provides a method for inverting the microphysical structure of layered cloud precipitation, a method for determining artificial rain enhancement potential, and a system. The method includes: acquiring raw power spectrum data from a micro-rain radar and observation data from a laser raindrop spectrometer; performing noise filtering, attenuation correction, and Doppler velocity aliasing removal on the power spectrum data; performing vertical airflow correction on the power spectrum based on raindrop spectrometer data and radar minimum reliable altitude data; identifying the phase states of precipitation particles at different altitudes based on the corrected data; and selectively inverting the raindrop spectrum distribution and microphysical characteristic parameters at each altitude layer according to the phase state. Beneficial effects: By comprehensively controlling quality and correcting vertical airflow, the reliability and accuracy of data inversion are significantly improved; a multi-parameter decision tree model is used to achieve fine phase state identification; and the combination of real-time microphysical parameters and historical statistical thresholds improves the accuracy of artificial rain enhancement potential determination and the efficiency of operational catalysis.

Description

Technical Field

[0001] This invention relates to the field of meteorology, specifically to a method for inverting the microphysical structure of layered cloud precipitation and a method and system for determining artificial rain enhancement potential. Background Technology

[0002] In the fields of meteorological observation and weather modification, the inversion of the microphysical structure of stratiform cloud precipitation is a key technology for improving the accuracy of precipitation forecasting and optimizing artificial rain enhancement operations. Currently, the observation of stratiform cloud precipitation mainly relies on remote sensing equipment, such as the MRR-2 micro-rain radar and the OTT ground-based laser raindrop spectrometer. The micro-rain radar is a vertically pointing Doppler radar that provides power spectrum data of precipitation particles in the vertical direction, including particle falling velocity and reflectivity information; the laser raindrop spectrometer measures the size and velocity distribution of ground precipitation particles through optical principles, providing direct microphysical parameters. However, existing technologies have significant limitations in inverting the microphysical structure of stratiform clouds.

[0003] First, the raw power spectrum data from light rain radar is susceptible to environmental interference, including noise, signal attenuation, and Doppler velocity aliasing. Noise mainly originates from atmospheric background clutter and the equipment itself, which can mask the true precipitation signal; signal attenuation is particularly pronounced under heavy precipitation or high humidity conditions, leading to an underestimation of reflectivity; and Doppler velocity aliasing, due to the limitations of the radar sampling frequency, causes ambiguity in velocity measurements. Existing quality control methods often employ simple filtering or thresholding, but fail to systematically integrate the influence of vertical airflow, resulting in insufficient data reliability.

[0004] Secondly, precipitation particle phase identification is the core of microphysics inversion, but existing methods mostly rely on a single parameter (such as the equivalent radar reflectivity factor) or empirical thresholds, resulting in low accuracy under complex precipitation conditions. Stratiform cloud precipitation is often accompanied by a bright band in the melt layer, where solid and liquid particles coexist and phase changes drastically. Existing techniques typically use a simple combination of reflectivity factor and velocity for classification, but this is insufficient to distinguish particle phases.

[0005] In assessing the potential for artificial rain enhancement, existing methods primarily rely on macroscopic meteorological data (such as cloud top temperature and cloud thickness) or simple radar echo characteristics, lacking detailed analysis of microphysical processes. Historical statistical thresholds are often extracted from long-term data but fail to adapt to real-time changes in precipitation microstructure, leading to inaccurate selection of operation timing and location, and low catalytic efficiency. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method for inverting the microphysical structure of layered cloud precipitation, a method and system for determining artificial rain enhancement potential.

[0007] This invention is achieved through the following technical solution:

[0008] Firstly, a method for inverting the microphysical structure of precipitation in layered clouds includes the following steps:

[0009] S1: Acquire raw power spectrum data from the micro-rain radar and observation data from the laser raindrop spectrometer;

[0010] S2: Perform quality control processing on the raw power spectrum data. The quality control processing includes at least noise filtering, attenuation correction, and Doppler velocity aliasing removal.

[0011] S3: Based on the observation data of the laser raindrop spectrometer and the power spectrum data of the light rain radar at the lowest reliable altitude, perform vertical airflow correction on the light rain radar power spectrum after quality control processing;

[0012] S4: Based on the power spectrum data corrected by vertical airflow, identify the phase state of precipitation particles at different heights within the detection range of the micro-rain radar;

[0013] S5: Based on the identified precipitation particle phase, selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

[0014] Preferably, step S3, which involves performing vertical airflow correction on the rain radar power spectrum after quality control processing based on the observation data from the laser rain spectrometer and the rain radar power spectrum data at the lowest reliable altitude, specifically includes:

[0015] S31: Filter out unreasonable data in the observation data of the laser raindrop spectrometer;

[0016] S32: Calculate the characteristic falling velocity v of the laser raindrop spectrometer. ott ;

[0017] S33: Calculate the characteristic fall velocity v of the light rain radar at the lowest reliable altitude. MRR ;

[0018] S34: If |v MRR -v ott If | is greater than a preset threshold, then use v ott Based on |v MRR -v ott | Values ​​are used to correct the power spectrum of the micro-rain radar for vertical airflow.

[0019] Preferably, step S4, which involves identifying the phase states of precipitation particles at different altitudes within the detection range of the micro-rain radar, specifically includes:

[0020] Based on the power spectrum data corrected for vertical airflow, the equivalent radar reflectivity factor Z for each altitude layer is calculated.e The average falling velocity of the particles, w; the Doppler spectral width, σ; the skewness of the velocity distribution, sk; and the kurtosis, kur.

[0021] Identify the bright band in the melt layer and determine the height BB at the top of the bright band. top And the height of the bottom of the bright strip BB Bottom ;

[0022] The above parameters are input into a pre-trained decision tree model for phase identification;

[0023] Based on the output of the decision tree model, precipitation particle phases are classified as: drizzle, rain, snow, mixed phase, and hail.

[0024] Preferably, the classification logic of the decision tree model is learned based on historical stratiform cloud precipitation observation data of the target area, including the following steps:

[0025] Based on the average falling velocity w, the Doppler spectral width σ, and the equivalent radar reflectivity factor Z e The calculated empirical value of the raindrop's falling speed v Rain And the empirical value v of the falling speed of ice crystal particles Snow To determine a preliminary decision branch;

[0026] The current distance to the height of the library is compared with the position of the bright band, and the preliminary judgment branch is mapped to a set of snow, mixed phase, or liquid / hail phase to be further subdivided based on the comparison result;

[0027] The logic for combining the falling speed feature and the spatial position feature includes one of the following cases:

[0028] When v Snow Within the range of w±σ and v Rain When the height is greater than w-σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is not lower than the bottom of the bright band, the mixed phase or snow is output.

[0029] Or, when v Snow and v Rain When all are within the range of w±σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is higher than the bottom of the bright band, the mixed phase or snow is output.

[0030] Or, when v Rain Within the range of w±σ and v Snow When the height is less than w-σ, if the height is below the top of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is above the top of the bright band, the mixed phase or snow is output.

[0031] Preferably, after mapping the preliminary judgment branch to snow, mixed phase, or a set of liquid / hail phases to be subdivided based on the comparison results, the liquid / hail phases to be subdivided are further refined, specifically as follows:

[0032] Using the skewness sk and the equivalent radar reflectivity factor Z of adjacent height layers e Differences and maximum diameters will be used to classify the liquid / hail phases to be subdivided into drizzle, rain, or hail.

[0033] If the maximum diameter of the particles is greater than a preset diameter threshold, the precipitation type is identified as hail; otherwise, it is identified as drizzle or rain.

[0034] When the skewness sk is less than or equal to a preset skewness threshold, and the equivalent radar reflectivity factor Z of the adjacent height layer... e When the difference is greater than or equal to a preset reflectance difference threshold, the liquid precipitation is classified as drizzle.

[0035] Otherwise, the liquid precipitation will be classified as rain.

[0036] Preferably, step S5, which involves selectively inverting and calculating the raindrop spectral distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phases, specifically includes:

[0037] When the precipitation particle phase is identified as drizzle or rain, the first calculation path is executed: based on the corrected power spectrum data, the raindrop spectrum distribution N(D,i) at that altitude layer is inverted, and the liquid water content LWC and radar reflectivity factor Z are calculated. e Rainfall intensity RR, mass-weighted average diameter D m and generalized intercept parameter N w ;

[0038] When the precipitation particle phase is identified as snow, the second calculation path is executed: based on the equivalent radar reflectivity factor Z. e Using Z e Based on the empirical relationship with snow intensity SR, calculate the snow intensity SR for this altitude level;

[0039] When the precipitation particle phase is identified as a mixed phase or hail, the first calculation path and the second calculation path are not executed.

[0040] Secondly, a method for determining the artificial rain enhancement potential of stratiform clouds includes:

[0041] Using the method described in any one of claims 1-6, the microphysical characteristic parameters of the target region at a specific time are obtained by inversion;

[0042] The equivalent radar reflectivity factor Z obtained by inversion, located below the melting layer, is...e The particle's falling velocity is used as a discrimination parameter;

[0043] The discrimination parameters are compared with a preset rain enhancement potential discrimination threshold;

[0044] Based on the comparison results, the rain enhancement potential of the target area is determined.

[0045] The preset rain enhancement potential discrimination thresholds, including weak broadcastability thresholds and strong broadcastability thresholds, are obtained through statistical analysis of historical stratiform cloud precipitation processes in the target area. The weak broadcastability threshold is calculated based on the minimum values ​​of the discrimination parameters during the early and late stages of historical precipitation processes. The strong broadcastability threshold is calculated based on the minimum values ​​of the discrimination parameters during the middle stage of historical precipitation processes.

[0046] Preferably, the method further includes: determining a recommended operation time window and a recommended operation height range based on statistical analysis of historical stratiform cloud precipitation processes;

[0047] The recommended operating height range is determined based on real-time detection data as a catalytic temperature window. The lower limit of this catalytic temperature window is the 0°C layer height of the ambient atmosphere, determined by the following method, and the upper limit is the radar equivalent reflectivity. Echo peak height:

[0048] The height of the top of the bright band identified by the light rain radar is BB. top An upward offset by a fixed distance is used as the lower limit of the catalytic temperature window; this fixed distance is obtained using an empirical formula; wherein the empirical formula is:

[0049]

[0050] in, This is the vertical distance between the top height of the melt layer and the height of the 0°C layer. The elevation of the light-rain radar. The latitude of the micro-rain radar. The average terminal velocity of solid water-condensed particles above 0°C, as measured by micro-rain radar. This represents the number concentration of solid water condensate particles above 0°C retrieved from a light rain radar. The atmospheric density at a depth of 0°C. , , , This is an empirical coefficient. This is the calibration constant;

[0051] Furthermore, within the catalytic temperature window, the region where the phase state identification result of the micro-rain radar is a mixed phase is determined as a core operating region rich in supercooled water. In this core operating region, an operating strategy of increasing the dosage of the layered cloud cold cloud catalyst is implemented to promote the growth of ice crystal sublimation in the berberine process.

[0052] Thirdly, a system for determining the artificial rain enhancement potential of stratiform clouds includes an inversion subsystem, wherein the inversion subsystem comprises:

[0053] The data acquisition module is used to acquire the raw power spectrum data of the micro-rain radar and the observation data of the laser raindrop spectrometer.

[0054] The quality control module is used to perform noise filtering, attenuation correction, and Doppler velocity aliasing removal on the original power spectrum data, and to perform vertical airflow correction on the power spectrum of the rain radar after quality control processing based on the observation data of the laser rain spectrometer and the power spectrum data of the rain radar at the lowest reliable height.

[0055] Based on the power spectrum data corrected by vertical airflow, the phase state of precipitation particles at different heights within the detection range of the micro-rain radar is identified;

[0056] The phase identification module identifies the phase of precipitation particles at different heights within the detection range of the micro-rain radar, based on the power spectrum data corrected for vertical airflow.

[0057] The parameter inversion module is used to selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phases; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

[0058] Preferably, it also includes a discrimination subsystem, the discrimination subsystem comprising:

[0059] The discrimination module is used to determine the equivalent radar reflectivity factor Z located below the melting layer. e The particle falling velocity is compared with the preset rain enhancement potential judgment threshold, and the rain enhancement potential judgment conclusion is output.

[0060] The beneficial effects of this invention are as follows: The technical solution provided by this invention can significantly improve the quality of light rain radar data. Through comprehensive quality control, it effectively reduces the impact of noise, attenuation, and velocity aliasing on the power spectrum, thereby enhancing the reliability and consistency of the data. The vertical airflow correction stage effectively eliminates the interference of airflow on particle falling velocity measurement, making the inverted particle motion state closer to the real situation, and significantly improving the accuracy of phase identification and microphysical parameter calculation. Using a decision tree model based on multi-parameter fusion for phase identification can adapt to the classification needs under complex precipitation conditions, significantly improving the distinction accuracy between drizzle, rain, snow, mixed phases, and hail, especially in the region near the phase-variable melting layer. The selective inversion mechanism based on phase identification results ensures a high degree of matching between the calculation of microphysical feature parameters and the physical properties of precipitation, making the liquid water content... The acquisition of key parameters such as precipitation, radar reflectivity factor, rainfall intensity, mass-weighted average diameter, generalized intercept parameter, and snow intensity is more reasonable and reliable. In terms of artificial rain enhancement potential assessment, by combining real-time inverted microphysical characteristic parameters with discrimination thresholds based on historical data statistics, the timing and location of rain enhancement can be assessed more accurately, significantly improving the scientific nature of operational decisions and catalytic efficiency. At the same time, by determining the recommended operational time window and catalytic temperature window, the catalyst delivery strategy is optimized, especially in the core operational area rich in supercooled water, where targeted catalysis is implemented to effectively promote the ice crystal sublimation growth process, thereby improving the overall effect of artificial rain enhancement. The modular design of the system realizes automatic data processing and discrimination functions, greatly improving the automation level of the inversion and discrimination process, facilitating integration into the existing meteorological observation network, and enhancing practicality and operability. Attached Figure Description

[0061] Figure 1 This is a flowchart of the method for inverting the microphysical structure of layered cloud precipitation in an embodiment of the present invention.

[0062] Figure 2 This is a technical roadmap of the layered cloud precipitation microphysical structure inversion method in this embodiment of the invention.

[0063] Figure 3 This is a diagram illustrating the noise level estimation process in an embodiment of the present invention.

[0064] Figure 4 The reflectance spectrum is obtained after noise filtering, conventional processing of MRR-2 micro-rain radar data, and defolding processing in the embodiments of the present invention.

[0065] Figure 5 This is a schematic diagram of the unfolding process in an embodiment of the present invention.

[0066] Figure 6 This is a schematic diagram of raindrop spectrum data quality control for a single sample in an embodiment of the present invention.

[0067] Figure 7 This is a flowchart of the first precipitation type estimation in an embodiment of the present invention.

[0068] Figure 8 This is a flowchart of the second precipitation type estimation and parameter inversion in an embodiment of the present invention. Detailed Implementation

[0069] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the embodiments of this invention will be further described in detail below with reference to the accompanying drawings. This invention provides a method for inverting the microphysical structure of layered cloud precipitation, a method and system for determining artificial rain enhancement potential, which can achieve accurate inversion of the microphysical structure of layered cloud precipitation and intelligent determination of artificial rain enhancement potential.

[0070] This embodiment provides a method for inverting the microphysical structure of layered cloud precipitation and a system for determining artificial rain enhancement potential. The method integrates observational data from micro-rain radar and a laser raindrop spectrometer, and through steps such as data quality control, vertical airflow correction, phase identification, and parameter inversion, achieves accurate inversion of the microphysical structure of precipitation at different altitudes. Based on this, the microphysical characteristic parameters obtained through inversion are used to determine the potential for artificial rain enhancement, including threshold comparison and operational strategy optimization. The system includes an inversion subsystem and a determination subsystem. The inversion subsystem is responsible for data acquisition, quality control, phase identification, and parameter inversion, while the determination subsystem outputs rain enhancement potential conclusions and operational recommendations based on the inversion results.

[0071] A study of the characteristics of stratiform cloud precipitation in a certain region revealed that stratiform cloud precipitation generally exhibits a bright band in the melt layer. Above the bright band, the precipitation mainly consists of solid precipitation particles such as snow, with a relatively slow terminal velocity. Near the bright band, the precipitation type is mixed-phase. Below the bright band, the precipitation type is mainly drizzle and rain, with significantly increased reflectivity and terminal velocity. The overall trend of precipitation microphysical parameters is: mid-precipitation > late-precipitation > early-precipitation. The average values ​​of precipitation microphysical parameters are greater on ridges than in valleys. During the mid-precipitation phase, rainfall intensity and liquid water content are significantly higher near the ground and near the melt layer. This is because supercooled water above 0℃ is abundant, capable of forming larger ice-phase particles, resulting in larger and more intense raindrops after melting. The average values ​​of rainfall intensity, radar reflectivity factor, and liquid water content show relatively small changes with altitude. The mass-weighted average diameter Dm gradually decreases with increasing altitude, while the generalized intercept parameter Nw based on liquid water content gradually increases, indicating that the precipitation process involves collisions and coalescence, resulting in larger scales and fewer particles. Based on the characteristics of stratiform cloud precipitation, the microphysical structure inversion method for stratiform cloud precipitation in this embodiment of the invention effectively improves data accuracy, and the resulting raindrop spectrum distribution and inversion physical parameters are more scientific and reasonable.

[0072] See Figure 1 As shown, a method for inverting the microphysical structure of layered cloud precipitation is presented, which includes the following steps:

[0073] S1: Acquire raw power spectrum data from the micro-rain radar and observation data from the laser raindrop spectrometer;

[0074] S2: Perform quality control processing on the raw power spectrum data. The quality control processing includes at least noise filtering, attenuation correction, and Doppler velocity aliasing removal.

[0075] S3: Based on the observation data of the laser raindrop spectrometer and the power spectrum data of the light rain radar at the lowest reliable altitude, perform vertical airflow correction on the light rain radar power spectrum after quality control processing;

[0076] S4: Based on the power spectrum data corrected by vertical airflow, identify the phase state of precipitation particles at different heights within the detection range of the micro-rain radar;

[0077] S5: Based on the identified precipitation particle phase, selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

[0078] For details, please refer to Figure 2 As shown, the rain radar used in this embodiment of the invention is the MRR-2 rain radar manufactured by METEK GmbH in Germany. It is a vertically pointing continuous frequency modulated wave Doppler radar. The MRR-2 compares the frequency changes (i.e., Doppler shift) of the scattered electromagnetic waves in the signal processor in the oscillator to obtain the terminal velocity of the falling precipitation particles. Then, based on the relationship between the raindrop diameter and the terminal velocity, the vertical distribution of the raindrop spectrum can be obtained. Subsequently, the radar reflectivity factor, rainfall intensity, liquid water content, mass-weighted average diameter, generalized intercept parameters and other profile information can be obtained by inverting the raindrop spectrum.

[0079] The laser rain spectrometer is the German OTT Parsivel2 model. Its working principle is mainly based on the attenuation effect of precipitation particles on the laser beam. It can emit a horizontal laser beam. When no precipitation particles fall through the laser beam, the output voltage is the maximum voltage received. When precipitation particles pass through the laser beam, the output voltage is proportional to the size of the particles that passed through. The falling speed of the precipitation particles is derived from the duration of the electronic signal. Based on the above determinants, information such as particle spectrum, precipitation type, rain intensity, and radar reflectivity can be calculated. It can measure precipitation weather phenomena such as drizzle, light rain, rain, sleet, snow, sleet, freezing rain, and hail.

[0080] The target area data in the embodiments of the present invention are all taken from the MRR-2 micro-rain radar and the OTT laser raindrop spectrometer deployed at the observation and research station in a certain region. The two instruments are deployed at a distance of less than 10m. The data are selected from multiple typical stratiform cloud rainfall processes throughout the year.

[0081] In step S2: the raw power spectrum data of the micro-rain radar is subjected to quality control to achieve noise removal, Doppler velocity aliasing removal, and attenuation correction for the micro-rain radar.

[0082] Noise Reduction: Besides the echo signal from the light-rain radar, unavoidable noise exists at the input of the radar receiver. The radar's data processing system calculates rainfall intensity based on the total received signal power. It cannot automatically distinguish between contributions from raindrops and contributions from noise. If the total signal is directly treated as entirely precipitation, the calculated rainfall intensity will naturally equal the actual rainfall intensity plus the noise-equivalent rainfall intensity. Even if this false rainfall rate is small, its persistence can lead to significant deviations in cumulative rainfall.

[0083] See Figure 3 As shown, the method for estimating the power spectrum noise level is as follows: In the first step, the ratio r1 = var1 / (mean1) is determined using the power spectrum variance var1 and mean1. 2 . When r1>n -1 If the highest spectral line in the power spectrum is not found, it is assumed that the power spectrum contains a signal peak (noise). The system then removes this highest spectral line, deeming it most likely a signal rather than noise. After removal, the system recalculates the variance, mean, and ratio r² based on the remaining spectral data, repeating this process until r² is reached. i ≤n -1 At this point, all remaining spectral lines with lower values ​​are unanimously considered pure noise, and the estimated noise level is equal to the mean. i Finally, the system will average the remaining spectral lines. i The estimated noise level is determined as the final noise level. Before all subsequent processing begins, this estimated noise level is subtracted from each spectral line of the original power spectrum to obtain a noise-removed signal data. Figure 3 In the process, the stopping condition is met after removing the 12th spectral line. Before further processing the power spectrum, the estimated power spectrum noise is subtracted from each spectral line.

[0084] De-Doppler velocity aliasing: The software built into the micro-rain radar makes two inherent assumptions when processing data that are inconsistent with reality: first, it assumes that all precipitation is liquid; second, it presupposes that the falling velocity of particles is only in the range of 0 to 12 m / s. However, in real stratiform cloud precipitation, solid particles (such as snowflakes) may move upward due to airflow (manifesting as negative velocity) or have a greater falling velocity. To correct this, we must extend the velocity detection range from the original 0–12 m / s (i.e., only downward falling velocity) to a more physically accurate range of −12–24 m / s, constructing a so-called "triple spectrum".

[0085] The system reads power spectrum data from the raw data file and, in conjunction with the radar's calibration parameters, first converts it into a quantity related to "reflectivity," which has a more defined physical meaning. This process can be understood as translating the raw, unprocessed electrical signal into a "reflectivity spectrum" that can represent the size and number of particles.

[0086] The reflectivity is calculated from the raw power spectrum f(n,i) received by the micro-rain radar, i.e.:

[0087] (1)

[0088] In the formula, i is the number of range bins (i=0,...,31), n ​​is the number of bins in the Doppler spectrum (n=0,...,63), f(n,i) data is stored in the original raw data file, TF(i) is the transfer function specific to each altitude, C is the radar calibration constant, and ∆h is the range resolution (unit: m). Reflectivity η(n,i) is the backscattering cross-section per unit volume, in m². -1 TF(i) and C are stored in the *.raw file.

[0089] Then, the reflectivity of the n Doppler spectra is converted to be related to the velocity v according to formula (2).

[0090] (2)

[0091] In the formula f sampling It is 125 kHz, n max For 64, i max Given 32 and λ as the wavelength of 1.24 cm, substituting these values ​​into formula (2) yields: The unit is m -2 s.

[0092] Based on the detection principle of the MRR-2 micro-rain radar, the unambiguous velocity range detected by the MRR-2 micro-rain radar is 0–12 m / s. By combining power spectrum information from adjacent (above and below) height levels to determine the falling velocity of water condensate particles, the original velocity range is extended from 0–12 m / s to −12–24 m / s.

[0093] Raw radar data can only "see" particles within a "velocity window" of 0-12 m / s. To see particles outside this window, we utilize the characteristic of precipitation being continuous in the vertical direction, meaning that the velocities of particles at adjacent altitudes are correlated.

[0094] See Figure 4 and Figure 5 As shown, the dealiasing velocity spectrum is generated by utilizing the vertical continuity of the velocity profile. The specific steps are as follows: Velocity range expansion: (1) For the j-th distance library, the falling velocity of the particle is determined by combining the power spectrum information of the adjacent height layers above and below; the velocity range of the (j-1)-th distance library is expanded to −12~0 m / s; the velocity range of the (j+1)-th distance library is expanded to 12~24 m / s. (2) Determine the peak point of the spectrum of the j-th distance library; determine the true spectrum (de-aliasing). (3) Find the peak point with the strongest energy from the original spectrum of the current library (j-th). Starting from this peak point, search upwards to find the value below the noise level between it and the next library (j-1). If it does not exist, determine the upper limit of the de-folded spectrum at the position of the minimum value, that is, determine the boundary of the true spectrum in the low-speed direction. (4) Similarly, search downwards from the peak point to determine the upper limit of the de-folded spectrum between it and the next library (j+1), that is, find the point where the signal strength is below the noise level, or the minimum value in that area. This position determines the boundary of the true spectrum in the high-speed direction. Through the above steps, the true, non-aliased velocity spectrum of the precipitation particles in the j-th layer is finally accurately cut out from the extended triple spectrum.

[0095] Attenuation correction: The intensity of radar waves decreases along their propagation path due to absorption and scattering by precipitation particles; this process is called attenuation. The attenuation effect is particularly pronounced in moderate to high-intensity rainfall. Without correction, reflectivity will be underestimated, leading to a systematic bias in the retrieved rainfall intensity.

[0096] First, the raindrop size distribution needs to be determined. Based on the power spectral reflectance related to velocity v obtained in the preceding steps... This allows us to establish the relationship between power spectrum reflectivity and raindrop diameter D.

[0097] (3)

[0098] This conversion needs to consider the effect of air density variations with altitude on the terminal velocity of falling particles. Therefore, a correction factor is introduced to adjust the falling velocity, namely:

[0099] (4)

[0100] The correction factor δv(h) is a function of the height h, specifically expressed as δv(h) = (1 + 3.68·10 -5 ·h+1.71·10 -9 ·h 2 Substituting into formula (3), we get:

[0101] (5)

[0102] Power spectrum reflectance after velocity correction Divide by the backscattering cross section σ of the raindrop particle with diameter D. b (Based on the Mie scattering theory), the desired raindrop spectral distribution can be obtained. It represents the number of raindrops per unit volume and unit diameter.

[0103] (6)

[0104] After determining the raindrop spectrum distribution Afterwards, attenuation correction can be performed. To distinguish them, we use the subscript "a" to mark variables that have not undergone attenuation correction. Path-Integrated Attenuation (PIA) is a key parameter, representing the total attenuation of the radar signal along its propagation path. The calculation uses range resolution Δr, range library number denoted by i, and single-event extinction coefficient. The diameter integration step size is defined as ΔD using Mie scattering theory. nn =(D nn+1 -D nn-1 ) / 2.

[0105] The attenuation correction calculation starts from the first distance library (i=1), assuming PIA(r0)=1, and is performed using an iterative algorithm:

[0106] Iterative calculation (for each distance library i, from 1 to the maximum number of libraries):

[0107] a. Calculate the preliminary corrected particle distribution:

[0108]

[0109] Here, PIA(r_{i-1}) is the path integral decay value of the previous distance library (for i=1, use PIA(r0)=1). This step performs preliminary decay compensation for the uncorrected particle distribution.

[0110] b. Calculate the initial attenuation coefficient:

[0111]

[0112] The initial attenuation coefficient of the current distance library is calculated by summing all diameter channels.

[0113] c. Calculate the final particle distribution:

[0114]

[0115] This formula is used to further correct the particle distribution to compensate for the attenuation effect.

[0116] d. Calculate the final attenuation coefficient:

[0117]

[0118] The attenuation coefficient is recalculated based on the corrected particle distribution.

[0119] e. Update Path Integral Decay (PIA):

[0120]

[0121] Update the PIA value to reflect the cumulative decay of the current distance library.

[0122] f. Exit condition check:

[0123] If PIA(ri) > 10, the calculation terminates. Because the algorithm is unstable at large decay values ​​(PIA > 10), it is only applicable to the case where PIA ≤ 10.

[0124] g. Iteration:

[0125] Set i = i + 1 and return to step a to process the next distance library.

[0126] Because the algorithm becomes unstable when the attenuation value is too large, attenuation correction is only applicable when the path integral attenuation (PIA) is ≤ 10. Given this limitation, the system provides both attenuation-corrected and non-attenuation-corrected inversion variables for users to choose from based on data quality and application scenarios.

[0127] In step S3: Based on the observation data from the laser raindrop spectrometer and the power spectrum data of the light rain radar at the lowest reliable altitude, vertical airflow correction is performed on the quality-controlled power spectrum of the light rain radar, specifically including:

[0128] S31: Filter out unreasonable data in the observation data of the laser raindrop spectrometer;

[0129] S32: Calculate the characteristic falling velocity v of the laser raindrop spectrometer. ott ;

[0130] S33: Calculate the characteristic fall velocity v of the light rain radar at the lowest reliable altitude. MRR ;

[0131] S34: If |v MRR -v ott If | is greater than a preset threshold, then use v ott Based on |v MRR -v ott | Values ​​are used to correct the power spectrum of the micro-rain radar for vertical airflow.

[0132] To more accurately retrieve the vertical distribution of raindrop spectra, we used observational data from a laser raindrop spectrometer (OTT Parsivel2) to estimate the vertical airflow influence on the micro-rain radar at the lowest reliable height and corrected the micro-rain radar data. The specific steps are as follows:

[0133] Quality control was performed on the OTT Parsivel2 data: Because the raindrop size in the first two scale channels (0.05–0.25 mm in diameter) of the OTT instrument is too small, it is easily affected by turbulence and ground splash, resulting in a low signal-to-noise ratio; therefore, this data was considered unusable. Additionally, raindrops larger than 8 mm in diameter are extremely rare in nature and were also excluded. Furthermore, to eliminate measurement errors (such as large raindrops being misjudged as high-speed small raindrops at the instrument edge, or large raindrops falling abnormally in strong winds), the empirical formula (7) for particle falling velocity versus diameter proposed by Atlas et al. was used. (See [reference]). Figure 6 As shown, unreasonable data that deviate from the empirical value by ±60% are filtered out.

[0134] (7)

[0135] in, Indicates diameter is The terminal velocity of falling precipitation particles, This represents the equivalent diameter of precipitation particles.

[0136] Vertical airflow correction: It is assumed that the vertical airflow variation of the micro-rain radar within its lowest reliable altitude (300 meters) is negligible, and that the vertical motion measured by the OTT ground-based laser raindrop spectrometer is close to zero. For each time interval (60 seconds), the power spectral reflectivity of the micro-rain radar is adjusted based on the OTT ground-based laser raindrop spectrometer data to ensure that the characteristic fall velocity calculated by the micro-rain radar is consistent with the results from the OTT ground-based laser raindrop spectrometer. The specific method is as follows:

[0137] For each time interval, the characteristic fall velocity v of the OTT is calculated using formula (8). ott .

[0138] (8)

[0139] Where, σ b The backscattering cross-section area at the frequency of the MRR-2 micro-rain radar m is the refractive index of water, which is 1.333. Liquid water at 24 GHz has a refractive index of |K|. 2 The value is approximately 0.92. Since the wavelength of the MRR-2 micro-rain radar is not small compared to the scale of raindrops in nature, Mie scattering theory is used here to calculate σ. b Replace D=(6V / π) 1 / 3 V is the volume of the raindrop.

[0140] The characteristic falling velocity v of light rain radar MRR The first moment of the power spectrum reflectivity is calculated, as shown in formula (9).

[0141] (9)

[0142] For each time interval (60s), the OTT ground laser raindrop spectrum is directly compared with the characteristic falling velocity v of the MRR-2 micro-rain radar at a height of 300m. MRR v calculated with OTT ott If the absolute difference between the two is |v MRR -v ott If the phase difference is greater than 0.2 m / s (this threshold corresponds to the velocity resolution of the micro-rain radar), then the power spectrum of the MRR is corrected for vertical airflow, with v ott For accuracy, the threshold of 0.2 m / s corresponds to the MRR velocity resolution.

[0143] In step S4: identifying the phase state of precipitation particles at different altitudes within the detection range of the micro-rain radar specifically includes:

[0144] Based on the power spectrum data corrected for vertical airflow, the equivalent radar reflectivity factor Z for each altitude layer is calculated. eThe average falling velocity of the particles, w; the Doppler spectral width, σ; the skewness of the velocity distribution, sk; and the kurtosis, kur.

[0145] Identify the bright band in the melt layer and determine the height BB at the top of the bright band. top And the height of the bottom of the bright strip BB Bottom ;

[0146] The above parameters are input into a pre-trained decision tree model for phase identification;

[0147] Based on the output of the decision tree model, precipitation particle phases are classified as: drizzle, rain, snow, mixed phase, and hail.

[0148] Specifically, the radar equivalent reflectivity Z is first calculated. e Doppler velocity, these two quantities, are calculated solely based on spectral reflectance, unaffected by the precipitation type of the hydrophobic particles, and from which the average falling velocity w, Doppler spectral width σ, skewness sk, and kurtosis kur of the hydrophobic particles are calculated. Radar equivalent reflectivity factor Z. e The Doppler spectrum reflects the size and quantity of precipitation particles; the Doppler spectral width σ describes the degree of difference in particle falling velocity; skewness reflects the symmetry of the cloud precipitation velocity distribution, and kurtosis measures the steepness of the velocity distribution. When there is a cloud-to-rain transition or a phase change in precipitation particles (such as snow melting into rain), the particle falling velocity will change, causing the velocity distribution to become no longer symmetrical and smooth. Skewness and kurtosis will gradually deviate from zero, preparing for subsequent differentiation of precipitation types.

[0149] Equivalent reflectivity factor Z e Calculated using formula (10):

[0150] (10)

[0151] Where λ is the radar wavelength and K is the dielectric constant of water. For velocity-dependent spectral reflectance;

[0152] The average falling velocity w of the hydrogel particles is calculated using formula (11):

[0153] (11)

[0154] in, Let be the spectral reflectance at the i-th distance and velocity v. The center velocity value of the i-th velocity channel;

[0155] The Doppler spectral width σ is calculated using formula (12):

[0156] (12)

[0157] in, The deviation between each speed channel and the average speed;

[0158] The skewness sk is calculated using formula (13):

[0159] (13)

[0160] Kurtivity is calculated using formula (14):

[0161] (14)

[0162] This project classifies hydrophobic particles at each altitude into five types: drizzle, rain, snow, mixed phase, and hail. This classification method is based on a decision tree model, comprehensively considering the empirical relationship between the falling velocity and equivalent radar reflectivity of different hydrophobic particles, the size characteristics of different hydrophobic particles, and the presence or absence of bright bands. First, it considers the relationship between radar reflectivity and rain velocity under the condition of bright echo bands, as studied by Atlas et al. Rain Or snowv Snow Empirical relationship of falling speed:

[0163] (15)

[0164] (16)

[0165] In addition, key geographical evidence is introduced: bright bands; a bright band detection scheme is adopted, and if a bright band exists, its top height (starting from the melt layer) is calculated. Top And the height of the bottom of the bright band (end of the melt layer) BB Bottom (Refer to Cha, 2009 method; specific existing techniques are not described here). Melting layer (0°C layer): Snowflakes begin to melt, and the exterior contains water, causing a sudden increase in radar reflectivity factor. This appears as a bright band on the vertical profile, which is the boundary between solid precipitation (snow) and liquid precipitation (rain).

[0166] See Figure 7 As shown, the decision tree for precipitation type estimation algorithm is mainly divided into three branches:

[0167] The classification logic of the decision tree model is learned based on historical stratiform cloud precipitation observation data of the target area, including the following steps: based on the average falling velocity w, the Doppler spectral width σ, and the equivalent radar reflectivity factor Z e The calculated empirical value of the raindrop's falling speed v Rain And the empirical value v of the falling speed of ice crystal particles SnowA preliminary judgment branch is determined; the current distance to the height of the library is compared with the position of the bright band, and the preliminary judgment branch is mapped to a set of snow, mixed phase, or liquid / hail phase to be further subdivided based on the comparison result;

[0168] The logic for combining the falling speed feature and the spatial position feature includes one of the following cases:

[0169] When v Snow Within the range of w±σ and v Rain When the height is greater than w-σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is not lower than the bottom of the bright band, the mixed phase or snow is output.

[0170] Or, when v Snow and v Rain When all are within the range of w±σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is higher than the bottom of the bright band, the mixed phase or snow is output.

[0171] Or, when v Rain Within the range of w±σ and v Snow When the height is less than w-σ, if the height is below the top of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is above the top of the bright band, the mixed phase or snow is output.

[0172] More specifically, the main branches of a decision tree are:

[0173] Branch A: v Snow Within the range of w±σ and v Rain When >w-σ,

[0174] Condition 1: Height < BB Bottom ,

[0175] Classified as: drizzle / rain, hail;

[0176] Condition 2: Height ≥ BB Bottom Or BB does not exist Bottom ,

[0177] Classified as: mixed phase,

[0178] Subcondition: If sk > -0.5 and w > v Snow ,

[0179] Classified as: Snow;

[0180] Branch B: When v Snow and v Rain When all are within the range of w±σ,

[0181] Condition 1: Height < BB BottomOr BB does not exist Bottom ,

[0182] Classified as: drizzle / rain, hail.

[0183] Condition 2: Height > BB Bottom ,

[0184] Classified as: mixed phase,

[0185] Subcondition: If sk > -0.5 and w > v Snow ,

[0186] Classified as: Snow;

[0187] Branch C: When v Rain Within the range of w±σ and v Snow When <w-σ,

[0188] Condition 1: Height < BB Top Or BB does not exist Top ,

[0189] Classified as: drizzle / rain, hail;

[0190] Condition 2: Height > BB Top

[0191] Classified as: Mixed phase

[0192] Subcondition: If sk > -0.5 and w > v Snow ,

[0193] It is classified as: snow.

[0194] The liquid / hail phases to be further subdivided are further refined as follows:

[0195] Using the skewness sk and the equivalent radar reflectivity factor Z of adjacent height layers e The differences and maximum diameter are used to classify the liquid / hail phases to be subdivided into drizzle, rain, or hail; when the maximum particle diameter is greater than a preset diameter threshold, the precipitation type is identified as hail, otherwise it is drizzle or rain; when the skewness sk is less than or equal to a preset skewness threshold, and the equivalent radar reflectivity factor Z of the adjacent height layer is... e If the difference is greater than or equal to a preset reflectance difference threshold, the liquid precipitation is classified as drizzle; otherwise, the liquid precipitation is classified as rain.

[0196] For details, please refer to Figure 8 As shown, since drizzle and rain consist only of droplet particles, this is determined by the skewness Sk of their falling velocity distribution and the radar reflectivity factor Z at various altitudes. e Difference ∆Z eTo distinguish them. Because drizzle consists of numerous, uniformly sized, and slowly falling tiny water droplets, the velocity distribution is concentrated at the low velocity end, resulting in a difference in reflectivity ∆Z. e A larger value indicates that significant evaporation occurs during the water droplet's descent at different altitudes; therefore, when skewness ≤ −0.5 and ∆Z e When the value is ≥1 dBZ, it is classified as drizzle; otherwise, it is classified as rain. Hail is defined according to the maximum diameter threshold of the Doppler velocity spectrum, which is set here to have a maximum diameter exceeding 5 mm.

[0197] In step S5: the selective inversion calculation of raindrop spectral distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phase states specifically includes:

[0198] When the precipitation particle phase is identified as drizzle or rain, the first calculation path is executed: based on the corrected power spectrum data, the raindrop spectrum distribution N(D,i) at this altitude layer is inverted, and the liquid water content LWC, radar reflectivity factor Z, rainfall intensity RR, and mass-weighted average diameter D are calculated. m and generalized intercept parameter N w When the precipitation particle phase is identified as snow, the second calculation path is executed: based on the equivalent radar reflectivity factor Z. e Using Z e Based on the empirical relationship with snow intensity SR, the snow intensity SR of the height layer is calculated; when the precipitation particle phase is identified as a mixed phase or hail, the first calculation path and the second calculation path are not executed.

[0199] Because the raindrop spectrum distribution is derived from the empirical relationship between the terminal velocity and diameter of condensate particles when using micro-rain radar, it does not output the raindrop spectrum distribution, reflectivity Z, rainfall intensity RR, liquid water content LWC, or mass-weighted average diameter D when the precipitation type is determined to be snow, mixed phase, or hail. m Generalized intercept parameter N w Rainfall parameters.

[0200] If the precipitation type is classified as snow, then its snow intensity can be calculated. Although solid precipitation varies more significantly than liquid precipitation, there is an empirical relationship between the equivalent radar reflectivity at different wavelengths and snow intensity. Snow intensity SR can be calculated by relating it to the equivalent radar reflectivity Z. e Empirical relationship calculation.

[0201] (17)

[0202] Where SR is in mm / h, Z e The unit is mm. 6 m -3 a and b are Z eThe coefficients relating to SR are a=56.00 and b=1.20 for the K-band.

[0203] Calculation process for liquid precipitation parameters (rainfall parameters for drizzle / rain):

[0204] If the precipitation type is liquid phase, then reflectivity Z, rainfall intensity RR, liquid water content LWC, and mass-weighted average diameter D can be calculated. m Generalized intercept parameter N w Iso-rainfall parameters. Based on the particle size distribution N(D,i) calculated using the previous formula (6) and the determined path integral attenuation PIA, the particle diameter distribution N is obtained through attenuation correction. a After (D,i), reflectivity factor Z, liquid water content LWC, rainfall intensity RR, and mass-weighted average diameter D are then calculated. m and the generalized intercept parameter N based on liquid water content w The calculation, where it is assumed that D0 equals D m .

[0205] The reflectance factor Z reflects the magnitude and concentration of precipitation, and is calculated using the following formula:

[0206] (18)

[0207] Liquid water content (LWC) indicates the total mass of liquid water in a unit volume of air, and is calculated using the following formula:

[0208] (19)

[0209] Rainfall intensity (RR) indicates the amount of rainfall per unit time, and the calculation formula is as follows:

[0210] (20)

[0211] Mass-weighted average diameter D m The formula for calculating the mass-weighted average raindrop diameter is as follows:

[0212] (twenty one)

[0213] Generalized intercept parameter N w The formula for calculating particle concentration is as follows:

[0214] (twenty two)

[0215] The above is an introduction to the joint inversion algorithm. This algorithm is implemented using the Python programming language, and the output file is a NetCDF file (standard meteorological data format). The specific output parameters, physical meanings, and units are shown in Table 1, and the time resolution is 1 minute.

[0216] Table 1. Introduction to the output parameters of the joint inversion algorithm

[0217] Parameter symbols Physical meaning unit W Falling speed <![CDATA[m·s -1 ]]> spectral width velocity spectrum <![CDATA[m·s -1 ]]> skewness skewness of velocity distribution / kurtosis kurtosis of velocity distribution / PIA Path integral decay of liquid water / PIA_all Excluding path integral decay of liquid water / Type Types of condensation (rain is marked as 10, drizzle as 5, snow as -10, mixture as -15, hail as -20) / LWC Liquid water content <![CDATA[g·m -3 ]]> RR Rain intensity <![CDATA[mm·h -1 <!-- 13 -->]]> SR Xueqiang <![CDATA[mm·h -1 ]]> Z Liquid water reflectivity dBZ Za Considering only the attenuation reflectivity of liquid water dBZ Ze Equivalent reflectance dBZ N(D) Droplet size distribution <![CDATA[log 10 (m -3 ·mm -1 )]]> <![CDATA[N w ]]> Intercept of the gamma distribution normalized to liquid water content <![CDATA[log 10 (m -3 ·mm -1 )]]> <![CDATA[D m ]]> Mass-weighted average diameter mm <![CDATA[BB bottom ]]> Bright strip bottom height m <![CDATA[BB top ]]> Bright band top height m

[0218] This invention also provides a method for determining the artificial rain enhancement potential of stratiform clouds, including:

[0219] Applying the aforementioned method for inverting the microphysical structure of layered cloud precipitation, the microphysical characteristic parameters of the target area at a specific time are obtained; the equivalent radar reflectivity factor Z, located below the melting layer, is then used as the inverted parameter. e The particle's falling velocity is used as a discrimination parameter;

[0220] The discrimination parameters are compared with a preset rain enhancement potential discrimination threshold;

[0221] Based on the comparison results, the rain enhancement potential of the target area is determined.

[0222] The preset rain enhancement potential discrimination thresholds, including weak broadcastability thresholds and strong broadcastability thresholds, are obtained through statistical analysis of historical stratiform cloud precipitation processes in the target area. The weak broadcastability threshold is calculated based on the minimum values ​​of the discrimination parameters during the early and late stages of historical precipitation processes. The strong broadcastability threshold is calculated based on the minimum values ​​of the discrimination parameters during the middle stage of historical precipitation processes.

[0223] Specifically, the working principle of assessing the artificial rainmaking potential of stratiform clouds is based on a deep understanding and quantitative analysis of the microphysical processes of stratiform cloud precipitation. Stratiform cloud precipitation typically has a relatively stable vertical structure, with the melt layer being a key height layer. Above this height, precipitation particles are mainly solid, while below it, they are mainly liquid. By analyzing the microphysical parameters below the melt layer, the potential for artificial rainmaking can be effectively determined.

[0224] Equivalent radar reflectivity factor Z e The calculation is based on the fundamental principles of radar meteorology. The radar reflectivity factor is defined as the sum of the sixth power of the diameters of all precipitation particles within a unit volume, with units of mm. 6 m -3 In practical applications, for ease of use, it is usually converted to logarithmic form dBZ. The calculation of the equivalent radar reflectivity factor not only considers the scattering characteristics of particles but also the attenuation effect along the propagation path, and the accuracy of the data is ensured through an attenuation correction algorithm.

[0225] The particle falling velocity is obtained using the Doppler radar velocities principle. Micro-rain radar calculates the falling velocity by measuring the Doppler frequency shift caused by precipitation particles. In actual observations, the particle falling velocity is affected by vertical airflow, therefore corrections are needed using observation data from a ground-based raindrop spectrometer. This combined correction method significantly improves the accuracy of falling velocity measurements.

[0226] The determination of the rain enhancement potential threshold is based on statistical analysis of a large amount of historical observation data. Through detailed analysis of multiple typical stratiform cloud precipitation events within a year, the distribution characteristics of microphysical parameters at different precipitation stages were statistically derived. The weak broadcastability threshold is derived based on the minimum parameter values ​​at the beginning and end of the precipitation phase, representing the basic conditions for stratiform cloud precipitation; the strong broadcastability threshold is derived based on the minimum parameter values ​​during the middle of the precipitation phase, representing the stronger state of stratiform cloud precipitation. This threshold determination method based on actual observation data ensures the scientific validity and practicality of the discrimination criteria.

[0227] In practice, the system calculates the microphysical parameters of the target area in real time and compares them with preset thresholds. When the measured parameters simultaneously meet all the threshold requirements for weak or strong playability, the system outputs the corresponding judgment conclusion. This multi-parameter joint judgment method avoids the uncertainty of single-parameter judgment and improves the reliability of the judgment results.

[0228] Taking a practical application in 2024 as an example, when the system detects the equivalent radar reflectivity factor Z below the melting layer of a certain meteorological station... e When the precipitation level reaches 15.3 dBZ, the particle falling velocity reaches 4.7 m / s, and the parameters above the melt layer also meet the corresponding conditions, the system automatically identifies it as a weakly sowing condition and outputs corresponding operational suggestions. Observational data after the actual operation showed a slow increase in the ground precipitation rate, verifying the correctness of the identification result.

[0229] This invention organically combines advanced observation technology, precise inversion algorithms, and scientific discrimination criteria to establish a complete method for judging the potential of artificial rain enhancement in stratiform clouds. This provides reliable technical support for weather modification operations and significantly improves the scientific nature and effectiveness of the operations.

[0230] Based on statistical analysis of historical stratiform cloud precipitation processes, a recommended operational time window and a recommended operational altitude range determined based on statistical characteristics of the melt layer height were identified. The recommended operational altitude range was determined as a catalytic temperature window region based on real-time detection data. The lower limit of this catalytic temperature window region is the 0°C layer height of the ambient atmosphere, determined using the following method, and the upper limit is the radar equivalent reflectivity Z. e Echo top height: The height of the top of the bright band identified by the light rain radar (BB) topAn upward offset by a fixed distance is used as the lower limit of the catalytic temperature window; this fixed distance is obtained using an empirical formula; wherein the empirical formula is:

[0231]

[0232] in, This is the vertical distance between the top height of the melt layer and the height of the 0°C layer. The elevation of the light-rain radar. The latitude of the micro-rain radar. The average terminal velocity of solid water-condensed particles above 0°C, as measured by micro-rain radar. This represents the number concentration of solid water condensate particles above 0°C retrieved from a light rain radar. The atmospheric density at a depth of 0°C. , , , This is an empirical coefficient. For calibration constants; and, within the catalytic temperature window, the region where the micro-rain radar phase identification result is a mixed phase is determined as a core operating region rich in supercooled water, and an operating strategy of increasing the dosage of layered cloud cold cloud catalyst is implemented in the core operating region to promote the growth of ice crystal sublimation in the bergelon process.

[0233] In-depth statistical analysis of historical stratiform cloud precipitation events in a certain region revealed a significant diurnal variation in the melt layer height, closely related to changes in atmospheric stratification stability caused by diurnal variations in solar radiation. Detailed studies of multiple typical stratiform cloud precipitation events in a given year showed that the melt layer height varied between 3300 and 5400 meters above sea level, with the highest frequency occurring around 4500 meters. Further analysis indicated that the peak frequency occurred primarily between 11:00 and 16:00 each day, with the frequency reaching its peak at 13:00. This statistical pattern is mainly attributed to increased atmospheric stratification instability due to enhanced afternoon solar radiation, leading to stronger updrafts and a higher melt layer height. Therefore, 11:00-16:00 was identified as the recommended operational window, as the supercooled water content within the stratiform clouds is relatively abundant during this period, and the ice-water conversion process is active, making it the most suitable time for artificial rain enhancement operations.

[0234] The recommended operating altitude range is determined based on a comprehensive analysis of the statistical characteristics of the melting layer height and real-time detection data. The lower limit of the catalytic temperature window is typically set at the 0°C layer height of the ambient atmosphere. This height is determined using the following method: The 0°C layer height is calculated by shifting the top of the bright band identified by the micro-rain radar upwards by a fixed distance. This fixed distance is calculated using an empirical formula that comprehensively considers multiple influencing factors, including the altitude and latitude of the micro-rain radar location, which directly affect the vertical distribution of air temperature; the average terminal velocity and particle number concentration of solid water condensate particles above 0°C measured by the micro-rain radar, which reflect the microphysical characteristics of solid precipitation particles; and the atmospheric density at the 0°C layer height, which affects the particle descent process and phase transition efficiency. The empirical coefficient is obtained through regression analysis of historical cases, while the calibration constant C is used to correct for the influence of local climate characteristics.

[0235] The empirical formula is:

[0236]

[0237] Particle type quantification index (atmospheric density is more stable in stratiform cloud environments, derived from altitude and latitude) (kg / m³) Particle type quantification indicators; the solid water condensate of stratiform cold clouds is mainly composed of snow crystals and small graupel, with a low proportion of mixed phase particles (Table 2). , , , These are empirical coefficients, see Table 3 for details. The calibration constant is (m).

[0238] Table 2 Classification of Particle Types in Stratiform Clouds

[0239] Particle type Classification criteria (wt, specific for stratiform clouds) Type value Physical meaning (stratified cloud scenario) Snow Crystal (Dominant Type) Wt∈(0.6,1.4) 75 Stratiform clouds are the most common type of cold clouds; they are low in density, melt slowly, and require a relatively long vertical path. Gramine / mixed phase Wt∈(1.4,2.0) 35 Stratiform clouds occurred during periods of heavy precipitation, with moderate density and a moderate melting path. Dense slug Wt∈[2.0,2.5] 10 Transitional scenes between stratiform and convective clouds: rapid melting and shortest path.

[0240] Table 3 Default values ​​for empirical coefficients

[0241] Particle type Classification criteria (wt, specific for stratiform clouds) Type value Physical meaning (stratified cloud scenario) Snow Crystal (Dominant Type) Wt∈(0.6,1.4) 75 Stratiform clouds are the most common type of cold clouds; they are low in density, melt slowly, and require a relatively long vertical path. Gramine / mixed phase Wt∈(1.4,2.0) 35 Stratiform clouds occurred during periods of heavy precipitation, with moderate density and a moderate melting path. Dense slug Wt∈[2.0,2.5] 10 Transitional scenes between stratiform and convective clouds: rapid melting and shortest path.

[0242] For example:

[0243] H_site = 2.848 # Radar altitude, in km

[0244] phi = 35 # Radar latitude, in °N

[0245] w_t = 2.5 # Average terminal velocity of the particle, in m / s

[0246] N_p = 200 # Particle number concentration, in m⁻³

[0247] rho_atm = 0.8 # Atmospheric density at 0°C, unit kg / m³

[0248] # Empirical coefficient (after optimization for layered cloud and cold cloud scenarios)

[0249] k1 = 0.28

[0250] k2 = 1.1

[0251] k3 = 150

[0252] k4 = 3.2

[0253] C = 50 # Calibration constant, which can be adjusted according to local data.

[0254] Particle type: Dense graupel

[0255] The calculated vertical distance between the top of the melt layer and the height of the 0°C layer is 518.82 m.

[0256] The upper limit of the catalytic temperature window is determined to be the radar equivalent reflectivity Z. e The echo top height represents the highest altitude at which a significant concentration of hydrophobic particles still exists within the cloud. In practical operations, this height information can be obtained in real time using the vertical detection capability of micro-rain radar, ensuring that the catalyst can diffuse and function effectively within the cloud. This method for determining the catalytic temperature window fully considers the internal thermodynamic structure and microphysical processes of the cloud, ensuring that the catalyst can function under the most suitable temperature conditions.

[0257] Within the defined catalytic temperature window, the core operating area is further determined using phase identification results from the micro-rain radar. When the phase identification result from the micro-rain radar shows a mixed phase, it indicates that both ice and water phase particles exist in the region, making it a typical supercooled water enrichment area. The theoretical basis for this criterion is the Bergieron process, which describes how, in ice-water mixed clouds, because the saturated vapor pressure at the ice surface is lower than that at the water surface, water vapor transfers from supercooled water droplets to ice crystals, promoting the sublimation and growth of ice crystals. The identification of mixed phase regions relies on multi-parameter comprehensive discrimination by the micro-rain radar, including a specific combination relationship between particle falling velocity and equivalent radar reflectivity, the skewness characteristics of the velocity spectrum, and the presence of obvious bright bands in the melt layer.

[0258] Within the mixed-phase area identified as the core operational zone, an operational strategy of increasing the dosage of stratified cloud cooling catalyst was implemented. This strategy is based on the principle of rapid transformation of supercooled water under the influence of artificial ice nuclei. When an appropriate amount of artificial ice nuclei is introduced into an area rich in supercooled water, the supercooled water will rapidly condense and grow on the ice nuclei, releasing latent heat, altering the thermal structure within the cloud, and promoting the formation and growth of precipitation particles. The purpose of increasing the catalyst dosage is to ensure the formation of a sufficient concentration of ice nuclei within the optimal catalytic temperature window, thereby maximizing the efficiency of the Begiron process and improving the efficiency of precipitation formation.

[0259] In practical operations, this method requires real-time micro-rain radar observation to dynamically adjust the operational altitude and catalyst dosage. For example, when a large altitude range of the mixed-phase region is detected, the catalyst dosage can be appropriately increased to ensure sufficient ice nucleus concentration throughout the effective area. Simultaneously, based on real-time observations of particle falling velocity and number concentration, operational parameters can be further optimized to improve the targeting and effectiveness of the catalytic operation. This precise operational method based on real-time detection significantly improves the scientific rigor and effectiveness of stratiform cloud artificial rain enhancement operations, providing strong technical support for water resource development and utilization.

[0260] This invention also provides a system for determining the artificial rain enhancement potential of stratiform clouds, including an inversion subsystem, the inversion subsystem comprising:

[0261] The data acquisition module is used to acquire the raw power spectrum data of the micro-rain radar and the observation data of the laser raindrop spectrometer.

[0262] The quality control module is used to perform noise filtering, attenuation correction, and Doppler velocity aliasing removal on the original power spectrum data, and to perform vertical airflow correction on the power spectrum of the rain radar after quality control processing based on the observation data of the laser rain spectrometer and the power spectrum data of the rain radar at the lowest reliable height.

[0263] Based on the power spectrum data corrected by vertical airflow, the phase state of precipitation particles at different heights within the detection range of the micro-rain radar is identified;

[0264] The phase identification module identifies the phase of precipitation particles at different heights within the detection range of the micro-rain radar, based on the power spectrum data corrected for vertical airflow.

[0265] The parameter inversion module is used to selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phases; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

[0266] The discrimination system also includes a discrimination subsystem, which includes:

[0267] The discrimination module is used to determine the equivalent radar reflectivity factor Z located below the melting layer. e The particle falling velocity is compared with the preset rain enhancement potential judgment threshold, and the rain enhancement potential judgment conclusion is output.

[0268] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for inverting the microphysical structure of precipitation in layered clouds, characterized in that, Includes the following steps: S1: Acquire raw power spectrum data from the micro-rain radar and observation data from the laser raindrop spectrometer; S2: Perform quality control processing on the raw power spectrum data. The quality control processing includes at least noise filtering, attenuation correction, and Doppler velocity aliasing removal. S3: Based on the observation data of the laser raindrop spectrometer and the power spectrum data of the light rain radar at the lowest reliable altitude, perform vertical airflow correction on the light rain radar power spectrum after quality control processing; S4: Based on the power spectrum data corrected by vertical airflow, identify the phase state of precipitation particles at different heights within the detection range of the micro-rain radar; S5: Based on the identified precipitation particle phase, selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

2. The method according to claim 1, characterized in that, Step S3, which involves performing vertical airflow correction on the micro-rain radar power spectrum after quality control processing based on the observation data from the laser raindrop spectrometer and the micro-rain radar power spectrum data at the lowest reliable altitude, specifically includes: S31: Filter out unreasonable data in the observation data of the laser raindrop spectrometer; S32: calculating a characteristic fall velocity v of the laser raindrop spectrometer ott ; S33: Calculate the characteristic fall velocity v of the light rain radar at the lowest reliable altitude. MRR ; S34: If |v MRR -v ott If | is greater than a preset threshold, then use v ott Based on |v MRR -v ott | Values ​​are used to correct the power spectrum of the micro-rain radar for vertical airflow.

3. The method according to claim 1 or 2, characterized in that, Step S4, which involves identifying the phase states of precipitation particles at different altitudes within the detection range of the micro-rain radar, specifically includes: Based on the power spectrum data corrected for vertical airflow, the equivalent radar reflectivity factor Z for each altitude layer is calculated. e The average falling velocity of the particles, w; the Doppler spectral width, σ; the skewness of the velocity distribution, sk; and the kurtosis, kur. Identify the bright band in the melt layer and determine the height BB at the top of the bright band. top And the height of the bottom of the bright strip BB Bottom ; The above parameters are input into a pre-trained decision tree model for phase identification; Based on the output of the decision tree model, precipitation particle phases are classified as: drizzle, rain, snow, mixed phase, and hail.

4. The method according to claim 3, characterized in that, The classification logic of the decision tree model is learned based on historical stratiform cloud precipitation observation data of the target area, and includes the following steps: Based on the average falling velocity w, the Doppler spectral width σ, and the equivalent radar reflectivity factor Z e The calculated empirical value of the raindrop's falling speed v Rain And the empirical value v of the falling speed of ice crystal particles Snow To determine a preliminary decision branch; The current distance to the height of the library is compared with the position of the bright band, and the preliminary judgment branch is mapped to a set of snow, mixed phase, or liquid / hail phase to be further subdivided based on the comparison result; The logic for combining the falling speed feature and the spatial position feature includes one of the following cases: When v Snow Within the range of w±σ and v Rain When the height is greater than w-σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is not lower than the bottom of the bright band, the mixed phase or snow is output. Or, when v Snow and v Rain When all are within the range of w±σ, if the height is lower than the bottom of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is higher than the bottom of the bright band, the mixed phase or snow is output. Or, when v Rain Within the range of w±σ and v Snow When the height is less than w-σ, if the height is below the top of the bright band, the set of liquid / hail phases to be subdivided is output; if the height is above the top of the bright band, the mixed phase or snow is output.

5. The method according to claim 4, characterized in that, After mapping the preliminary judgment branch to snow, mixed phase, or a set of liquid / hail phases to be further subdivided based on the comparison results, the liquid / hail phases to be further subdivided are further refined, specifically as follows: Using the skewness sk and the equivalent radar reflectivity factor Z of adjacent height layers e Differences and maximum diameters will be used to classify the liquid / hail phases to be subdivided into drizzle, rain, or hail. If the maximum diameter of the particles is greater than a preset diameter threshold, the precipitation type is identified as hail; otherwise, it is identified as drizzle or rain. When the skewness sk is less than or equal to a preset skewness threshold, and the equivalent radar reflectivity factor Z of the adjacent height layer... e When the difference is greater than or equal to a preset reflectance difference threshold, the liquid precipitation is classified as drizzle. Otherwise, the liquid precipitation will be classified as rain.

6. The method according to claim 3, characterized in that, Step S5, which involves selectively inverting and calculating the raindrop spectral distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phases, specifically includes: When the precipitation particle phase is identified as drizzle or rain, the first calculation path is executed: based on the corrected power spectrum data, the raindrop spectrum distribution N(D,i) at this altitude layer is inverted, and the liquid water content LWC, radar reflectivity factor Z, rainfall intensity RR, and mass-weighted average diameter D are calculated. m and generalized intercept parameter N w ; When the precipitation particle phase is identified as snow, the second calculation path is executed: based on the equivalent radar reflectivity factor Z. e Using Z e Based on the empirical relationship with snow intensity SR, calculate the snow intensity SR for this altitude level; When the precipitation particle phase is identified as a mixed phase or hail, the first calculation path and the second calculation path are not executed.

7. A method for determining the artificial rain enhancement potential of stratiform clouds, characterized in that, include: Using the method described in any one of claims 1-6, the microphysical characteristic parameters of the target region at a specific time are obtained by inversion; The equivalent radar reflectivity factor Z obtained by inversion, located below the melting layer, is... e The particle's falling velocity is used as a discrimination parameter; The discrimination parameters are compared with a preset rain enhancement potential discrimination threshold; Based on the comparison results, the rain enhancement potential of the target area is determined. The preset rain enhancement potential discrimination threshold, including the weak broadcastable condition threshold and the strong broadcastable condition threshold, is obtained by statistical analysis of historical stratiform cloud precipitation processes in the target area; the weak broadcastable condition threshold is statistically obtained based on the minimum values ​​of the discrimination parameters in the early and late stages of historical precipitation processes. The threshold for strong broadcastability is statistically derived based on the minimum value of the discrimination parameter during the mid-stage of historical rainfall.

8. The method according to claim 7, characterized in that, Also includes: Based on statistical analysis of historical stratiform cloud precipitation processes, a recommended operation time window and a recommended operation height range determined based on the statistical characteristics of the melt layer height were identified. The recommended operating height range is determined based on real-time detection data as a catalytic temperature window. The lower limit of this catalytic temperature window is the 0°C layer height of the ambient atmosphere, determined by the following method, and the upper limit is the radar equivalent reflectivity Z. e Echo peak height: The height of the top of the bright band identified by the light rain radar is BB. top An upward offset by a fixed distance is used as the lower limit of the catalytic temperature window; this fixed distance is obtained using an empirical formula; wherein the empirical formula is: in, This is the vertical distance between the top height of the melt layer and the height of the 0°C layer. The elevation of the light-rain radar. The latitude of the micro-rain radar. The average terminal velocity of solid water-condensed particles above 0°C, as measured by micro-rain radar. This represents the number concentration of solid water condensate particles above 0°C retrieved from a light rain radar. The atmospheric density at a depth of 0°C. , , , This is an empirical coefficient. This is the calibration constant; Furthermore, within the catalytic temperature window, the region where the phase state identification result of the micro-rain radar is a mixed phase is determined as a core operating region rich in supercooled water. In this core operating region, an operating strategy of increasing the dosage of the layered cloud cold cloud catalyst is implemented to promote the growth of ice crystal sublimation in the berberine process.

9. A system for determining the artificial rainmaking potential of stratiform clouds, characterized in that, Includes an inversion subsystem, the inversion subsystem comprising: The data acquisition module is used to acquire the raw power spectrum data of the micro-rain radar and the observation data of the laser raindrop spectrometer. The quality control module is used to perform noise filtering, attenuation correction, and Doppler velocity aliasing removal on the original power spectrum data, and to perform vertical airflow correction on the power spectrum of the rain radar after quality control processing based on the observation data of the laser rain spectrometer and the power spectrum data of the rain radar at the lowest reliable height. Based on the power spectrum data corrected by vertical airflow, the phase state of precipitation particles at different heights within the detection range of the micro-rain radar is identified. The phase identification module identifies the phase of precipitation particles at different heights within the detection range of the micro-rain radar, based on power spectrum data corrected for vertical airflow. The parameter inversion module is used to selectively invert and calculate the raindrop spectrum distribution and one or more microphysical characteristic parameters at different altitudes based on the identified precipitation particle phases; wherein, the microphysical characteristic parameters include parameters related to liquid precipitation and parameters related to solid precipitation.

10. A system for determining the artificial rainmaking potential of stratiform clouds, characterized in that, The system includes a discrimination subsystem, which comprises: The discrimination module is used to determine the equivalent radar reflectivity factor Z located below the melting layer. e The particle falling velocity is compared with the preset rain enhancement potential judgment threshold, and the rain enhancement potential judgment conclusion is output.

Images