A snowfall amount inversion method based on millimeter wave cloud radar and microwave radiometer

Abstract

The application discloses a snowfall inversion method based on a millimeter wave cloud radar and a microwave radiometer, and relates to the technical field of radar snowfall inversion, and comprises the following steps: S1, data input: millimeter wave cloud radar data and microwave radiometer data input; S2, time-space dimension matching of multi-source equipment data; S3, polar cold particle phase state classification, and elimination of non-snow phase state interference; S4, inversion of snowfall rates of different height layers; S5, cumulative snowfall amount obtained through time integration of whole-layer snowfall rates; and S6, result output. The application fully gives play to the complementary advantages of high time-space resolution of the millimeter wave cloud radar and rich atmospheric environmental parameters of the microwave radiometer, solves inherent defects of single-device inversion, greatly improves the precision and reliability of snowfall inversion in a polar cold environment, and provides data support for practical applications such as meteorological monitoring and early warning, snowstorm disaster prevention and disaster reduction in a polar cold region.

Description

Technical Field

[0001] This invention relates to the field of radar snowfall inversion technology, and more specifically to a snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer. Background Technology

[0002] Snowfall is a typical meteorological process in cold regions and winter. Accurate snowfall inversion is a core foundation for meteorological monitoring and early warning, blizzard disaster prevention and mitigation, road traffic safety assurance, and cold region water resource assessment. Snowfall inversion under extremely cold environments is a significant research direction in the field of atmospheric sounding technology, as its accuracy directly determines the scientific validity of winter meteorological decisions and the practical effectiveness of various application scenarios. Existing patents disclose the following technologies:

[0003] The patent with publication number CN118837978A and titled "Snowfall Calculation Method, Device, Equipment and Storage Medium" discloses the following: A snowfall calculation method, device, equipment and storage medium acquires a set of geographic environmental parameters and snowfall-related parameter information of the target monitoring area, constructs a geographic parameter feature map using the geographic environmental parameter set, generates several migration reference areas in the geographic environment database, constructs an independent reference snowfall prediction model based on the historical snowfall information of each migration reference area, and then performs model migration based on model weight mapping on the reference snowfall prediction model, and adjusts the migration strategy based on the model optimization period to achieve model migration and optimization between the migration reference area and the target monitoring area. On the one hand, this ensures the accuracy of snowfall calculation in monitoring areas under data-scarce scenarios, and on the other hand, it improves the adaptability of the snowfall prediction model to the target monitoring area, avoiding model incompatibility problems caused during model migration, and has good scenario adaptability.

[0004] Patent publication number CN105572764A, entitled "A Video-Based Snowfall Measurement Method," discloses the following: A video-based snowfall measurement method includes the following steps: acquiring snowfall images using a camera and capturing images using computer snapshot technology; collecting snowfall data using a snow chamber, with the chamber height serving as the image calibration dimension; enhancing image grayscale by using a snow chamber with a back color different from white; calibrating using an image window, with the window height aligned with the snow chamber; performing image grayscale comparison to distinguish between drifting snow and accumulated snow in each row of the image based on the density of background color and white dots; calculating the total snow accumulation based on the snow accumulation image height comparison calibration dimension; and calculating the real-time snowfall based on the increase in snow accumulation height over the image acquisition period. This invention is convenient, accurate, and enables continuous monitoring and snowfall monitoring in harsh environments. Furthermore, it has low equipment requirements and is easy to implement.

[0005] The aforementioned patents and existing snowfall inversion technologies are mostly based on single observation equipment. While millimeter-wave cloud radar can capture the microphysical characteristics of snowfall particles, it lacks environmental background information such as atmospheric temperature and humidity. Microwave radiation can obtain atmospheric vertical profile parameters, but its spatial resolution is low. Furthermore, the spatiotemporal resolution of the observation data from the two types of equipment is significantly heterogeneous, and there is no scientifically unified spatiotemporal matching rule. At the same time, traditional methods do not conduct accurate phase classification of snowfall particles for extremely cold environments, and are easily affected by non-snowfall particles, causing inversion bias. Overall, the inversion results have problems of large errors and low accuracy, making it difficult to meet the needs of practical applications. Summary of the Invention

[0006] The purpose of this invention is to provide a method for snowfall inversion based on millimeter-wave cloud radar and microwave radiometer in order to solve the above-mentioned technical problems.

[0007] To achieve the above objectives, the present invention specifically adopts the following technical solution:

[0008] This invention provides a method for retrieving snowfall based on millimeter-wave cloud radar and microwave radiometer, comprising the following steps:

[0009] S1. Data Input: Input millimeter-wave cloud radar data and microwave radiometer data into the data processing module;

[0010] S2. Multi-source data spatiotemporal dimension matching: The data processing module sequentially matches and aligns millimeter-wave cloud radar data with microwave radiometer data from both time and space dimensions.

[0011] S3. Extreme cold particle phase classification: By collecting observation samples of continuous snowfall processes in extremely cold regions, a parameter sample set of various extreme cold particle phases is constructed. The initial discrimination threshold of each parameter is determined based on statistical analysis, and the final threshold is determined through actual measurement, calibration and verification to achieve extreme cold particle phase classification.

[0012] S4. Inverting the snowfall rate at different altitude levels: First, calculate the instantaneous snowfall intensity at different altitude levels for each layer, and then obtain the snowfall rate of the entire layer through vertical weighted fusion;

[0013] S5. Calculate the total cumulative snowfall over the period: The cumulative snowfall is obtained by integrating the whole-layer snowfall rate obtained in step S4 over time.

[0014] S6. Output the results.

[0015] Specifically, snowfall rate refers to the amount of snowfall per unit time and per unit horizontal area, used to characterize the intensity of snowfall at a given observation time, and is measured in units of... Cumulative snowfall is the time integral of the snowfall rate over a certain period, and the unit is... ;

[0016] In one implementation, in step S2, the time resolutions of the cloud radar data and the microwave radiometer data are different in the time dimension; typically, the cloud radar data is recorded every 1 minute, while the microwave radiometer data is recorded every 10 seconds. Given that atmospheric temperature and humidity change relatively gradually over time, the time resolution of the cloud radar data is chosen as the time reference for matching. The specific operation is as follows:

[0017] For each cloud radar time Microwave radiation time with the closest matching time , ,like If the two sets of data match successfully, then the two sets of data are considered to be a match.

[0018] In one implementation, in step S2, the spatial difference between cloud radar data and microwave radiometer data is more significant. Although both cloud radar data and microwave radiometer data are vertical profile data, the cloud radar range database can contain hundreds to thousands of data points with a spatial resolution of 30 meters; while the microwave radiometer provides fixed data from 83 layers, uniformly covering a height range of 0 to 1000 meters. Since the spatial resolution of cloud radar data is higher than that of microwave radiometer data, the specific operation is based on the spatial resolution of cloud radar data as follows:

[0019] The altitude corresponding to each distance in the cloud radar database is used as a sub-element, and the data are sorted in ascending order to generate a baseline altitude profile. ,

[0020] ;

[0021] In the formula, As the reference height profile, These represent the heights corresponding to each distance in the cloud radar database, such as: Indicates the first The height corresponding to each library.

[0022] The microwave radiometer data was then interpolated to the reference height profile. Above, for any microwave radiometer variable At the reference height The interpolation result at point is:

[0023] ;

[0024] In the formula, , Distance in microwave radiometer data The two most recent effective observation altitudes meet the requirements. ; For height Microwave radiometer observations For height Microwave radiometer observations For height Interpolation results of the microwave radiometer.

[0025] In one implementation, in step S3, in order to eliminate the error caused by non-snowfall particles in snowfall rate inversion, particle phase classification is performed before snowfall amount inversion. Since the application scenario of snowfall amount inversion is extreme cold snowfall weather, extreme cold particle phase classification is only carried out for relevant phases in extreme cold environments. By collecting observation samples of continuous snowfall processes in extreme cold regions, a parameter sample set of various extreme cold particle phases is constructed. The initial discrimination threshold of each parameter is determined based on statistical analysis, and the final threshold is determined by actual measurement calibration and verification.

[0026] In one implementation, based on the inherent differences in the physical structure, motion characteristics, and environmental conditions of different particles, the reflectivity in wave cloud radar is used. radial velocity Velocity spectral width and atmospheric temperature in microwave radiometers and relative humidity As a basis for judgment, the system uses multi-condition combined logic to distinguish the five extreme cold particle phases: ice crystals, dry snow, graupel, mixed phase, and wet snow.

[0027] In one implementation, the specific steps for distinguishing the five types of extremely cold particle phase states are as follows:

[0028] S31. Definition of parameter sample set: For the first... Phase-like ( (corresponding to ice crystals, dry snow, graupel, mixed phase, and wet snow, respectively) for collection. Construct a parameter sample set from 100 valid observation samples:

[0029] In the formula, Reflectivity in cloud radar For radial velocity in cloud radar, For the velocity spectrum width in cloud radar, The atmospheric temperature is measured in the microwave radiometer. The relative humidity in the microwave radiometer;

[0030] S32. Calculation of the mean of the sample set:

[0031] ;

[0032] S33. Calculation of standard deviation of sample set:

[0033] ;

[0034] S34. Determination of the discrimination threshold: Based on the observation characteristics of extreme cold snowfall, take... This serves as the basic interval for the corresponding phase state parameters, i.e., the initial threshold.

[0035] S35. Repeat steps 31 to S34 to determine the initial threshold of each parameter, and complete the discrimination of the threshold range through independent observation sample verification and physical rationality verification.

[0036] In one implementation, in step S35, for the initial threshold of each parameter, the threshold interval discrimination conditions are determined through independent observation sample verification and physical rationality verification as follows:

[0037] For ice crystals, the threshold range is as follows: ;

[0038] For dry snow, the threshold range is as follows:

[0039] For graupel, the threshold range is as follows: ;

[0040] For mixed phases, the threshold range is as follows: ;

[0041] For wet snow, the threshold range is as follows: .

[0042] In one implementation, the specific operation of inverting the instantaneous snowfall rate in a single time interval in step S4 is as follows:

[0043] S41. Traverse each distance database along the cloud radar profile and calculate the instantaneous snowfall intensity at different altitudes. During traversal, first check the particle phase classification results of the current distance database; only calculate the snowfall rate when the phase is wet snow or dry snow. The calculation method adopts... Power-law relationship The relationship is transformed into:

[0044] ;

[0045] In the formula, The current distance corresponds to the cloud radar reflectivity. This refers to the intensity of a single layer of instantaneous snowfall. , These are empirical coefficients corresponding to different phase states, typically for dry snow. wet snow It can be localized according to the geographical features and snowfall characteristics of different extremely cold regions;

[0046] S42. The instantaneous snowfall intensity of a single layer is the hydrophobicity formation rate of snow particles within that layer, while the ground snowfall rate is the fusion result of snow particles from different altitude layers after falling to the ground. Therefore, a vertical weighted fusion of instantaneous snowfall from each layer is necessary. High-altitude snow particles are prone to sublimation and evaporation during their descent, making it difficult for them to reach the ground; low-altitude snow particles are more physically stable and more likely to reach the ground. Therefore, different weights are assigned to different altitude layers, with a higher weight for lower altitudes.

[0047] ;

[0048] In the formula, For the entire layer of snowfall rate, For the first The instantaneous snowfall intensity at a distance from the reservoir. For the first The height contribution weight of each distance library For the first The height of the distance to the warehouse, The maximum height of the effective snow layer. Ground height; the effective snow layer refers to the vertical height interval defined by the distances of the first and last effectively identified as dry or wet snow within the current vertical profile.

[0049] In one implementation, the specific method for accumulating the total snowfall over the time period in step S5 is as follows:

[0050] If the cloud radar observation time series is The corresponding total snowfall rate Then the time period Cumulative snowfall Cumulative snowfall The calculation formula is:

[0051] .

[0052] In one implementation, the cumulative snowfall is calculated via direct time integration when there is no missing data; if there is missing data, it is calculated using discrete integration to calculate the cumulative snowfall. The calculation formula is:

[0053] ;

[0054] In the formula, For the first Each observation time, For the first Each observation time, , for the first The length of each time period; for Snowfall rate per period for The snowfall rate per time period, where N is the total number of valid observations within the target time period. This represents the cumulative snowfall during the specified period.

[0055] The beneficial effects of this invention are as follows:

[0056] This invention fully integrates the complementary advantages of high spatiotemporal resolution of millimeter-wave cloud radar and rich environmental parameters such as atmospheric temperature and humidity from microwave radiometers. It addresses the inherent limitations of single-observation-device inversion from the data source, constructing a comprehensive, high-precision snowfall inversion system for extremely cold environments. By establishing spatiotemporal matching rules based on cloud radar, accurate alignment of multi-source data is achieved. Relying on the joint discrimination of multiple parameters from cloud radar and microwave radiometers, accurate classification of five particle phases in extremely cold environments is achieved, providing a core basis for effectively eliminating interference from non-snowfall phases. Furthermore, this invention employs a layered snowfall intensity calculation and vertical weighted fusion strategy, fully considering the physical characteristics of high-altitude snow particles easily sublimating and low-altitude particles more easily settling, making the inversion results more consistent with actual ground snowfall conditions. Accumulated snowfall is calculated through time integration, compatible with scenarios with missing data, significantly improving the robustness and applicability of the inversion method. This invention improves the accuracy, stability, and reliability of snowfall inversion in extremely cold environments, and can provide accurate and reliable data support for practical applications such as meteorological monitoring and early warning in extremely cold regions, disaster prevention and mitigation of blizzards, road traffic safety assurance, and scientific assessment of water resources in cold regions. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0058] Figure 1 This is a flowchart of a snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to the present invention.

[0059] Figure 2 This is a schematic diagram before microwave radiometer data interpolation.

[0060] Figure 3 This is a schematic diagram after interpolation of microwave radiometer data.

[0061] Figure 4 It is a cloud radar reflectivity map for a certain period of time.

[0062] Figure 5 It is a diagram showing the particle phase state results over a certain period of time.

[0063] Figure 6This is a cloud radar reflectivity map for another time period.

[0064] Figure 7 This is a diagram showing the particle phase state results for another time period.

[0065] Figure 8 It is a time series graph of snowfall rate over a certain period of time.

[0066] Figure 9 This is a time series graph of snowfall rate over another period.

[0067] Figure 10 It is a time series graph of cumulative snowfall over a certain period of time.

[0068] Figure 11 This is a time series graph of cumulative snowfall over another period. Detailed Implementation

[0069] To make the technical problems, technical solutions, and technical effects of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0070] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0071] Example 1

[0072] like Figure 1 As shown, this embodiment provides a method for retrieving snowfall based on millimeter-wave cloud radar and microwave radiometer, including the following steps:

[0073] S1. Data Input: Input millimeter-wave cloud radar data and microwave radiometer data into the data processing module;

[0074] S2. Multi-source data spatiotemporal dimension matching: The data processing module sequentially matches and aligns millimeter-wave cloud radar data with microwave radiometer data from both time and space dimensions.

[0075] S3. Extreme cold particle phase classification: By collecting observation samples of continuous snowfall processes in extremely cold regions, a parameter sample set of various extreme cold particle phases is constructed. The initial discrimination threshold of each parameter is determined based on statistical analysis, and the final threshold is determined through actual measurement, calibration and verification to achieve extreme cold particle phase classification.

[0076] S4. Inverting the snowfall rate at different altitude levels: First, calculate the instantaneous snowfall intensity at different altitude levels for each layer, and then obtain the snowfall rate of the entire layer through vertical weighted fusion;

[0077] S5. Calculate the total cumulative snowfall over the period: The cumulative snowfall is obtained by integrating the whole-layer snowfall rate obtained in step S4 over time.

[0078] S6. Output the results.

[0079] Specifically, snowfall rate refers to the amount of snowfall per unit time and per unit horizontal area, used to characterize the intensity of snowfall at a given observation time, and is measured in units of... Cumulative snowfall is the time integral of the snowfall rate over a certain period, and the unit is... ;

[0080] Example 2

[0081] like Figures 1 to 11 As shown, this embodiment provides a method for retrieving snowfall based on millimeter-wave cloud radar and microwave radiometer, including the following steps:

[0082] S1, millimeter-wave cloud radar data and microwave radiometer data input;

[0083] S2. Spatiotemporal data matching from multiple sources: Matching and aligning millimeter-wave cloud radar data with microwave radiometer data sequentially from both time and space dimensions.

[0084] In terms of time, cloud radar data and microwave radiometer data have different time resolutions; typically, cloud radar data is transmitted once every 1 minute, while microwave radiometer data is transmitted once every 10 seconds. Given that atmospheric temperature and humidity change relatively gradually over time, the time resolution of the cloud radar data is chosen as the time reference for matching. The specific operation is as follows:

[0085] For each cloud radar time Microwave radiation time with the closest matching time , ,like If the two sets of data match successfully, then the two sets of data are considered to be a match.

[0086] In terms of spatial dimension, the differences between cloud radar data and microwave radiometer data are more significant. Although both cloud radar and microwave radiometer data are vertical profile data, cloud radar range databases can contain hundreds to thousands of data points with a spatial resolution of 30 meters; while microwave radiometer data consists of a fixed 83 layers, uniformly covering the height range of 0 to 1000 meters. Since cloud radar data has a higher spatial resolution than microwave radiometer data, the specific operation is based on the spatial resolution of cloud radar data as follows:

[0087] The altitude corresponding to each distance in the cloud radar database is used as a sub-element, and the data are sorted in ascending order to generate a baseline altitude profile. ,

[0088] ;

[0089] In the formula, As the reference height profile, These represent the heights corresponding to each distance in the cloud radar database, such as: Indicates the first The height corresponding to each library.

[0090] The microwave radiometer data was then interpolated to the reference height profile. Above, for any microwave radiometer variable At the reference height The interpolation result at point is:

[0091] ;

[0092] In the formula, , Distance in microwave radiometer data The two most recent effective observation altitudes meet the requirements. ; For height Microwave radiometer observations For height Microwave radiometer observations For height Interpolation results of the microwave radiometer.

[0093] S3. Extreme cold particle phase classification to eliminate interference from non-snowfall phases: To eliminate errors caused by non-snowfall particles in snowfall rate inversion, particle phase classification was performed before snowfall amount inversion. Since the application scenario of snowfall amount inversion is extreme cold snowfall weather, extreme cold particle phase classification is only carried out for relevant phases under extreme cold environments. By collecting observation samples of continuous snowfall processes in extreme cold regions, parameter sample sets of various extreme cold particle phases were constructed. The initial discrimination thresholds of each parameter were determined based on statistical analysis, and the final thresholds were determined through actual measurement calibration and verification.

[0094] Based on the inherent differences in physical structure, motion characteristics, and environmental conditions of different particles, the reflectivity in wave cloud radar is used as an example. radial velocity Velocity spectral width and atmospheric temperature in microwave radiometers and relative humidity As a criterion, a multi-condition joint logic judgment is used to distinguish the five extreme cold particle phases: ice crystals, dry snow, graupel, mixed phase, and wet snow. The specific steps for distinguishing the five extreme cold particle phases are as follows:

[0095] S31. Definition of parameter sample set: For the first... Phase-like ( (corresponding to ice crystals, dry snow, graupel, mixed phase, and wet snow, respectively) for collection. Construct a parameter sample set from 100 valid observation samples:

[0096] In the formula, Reflectivity in cloud radar For radial velocity in cloud radar, For the velocity spectrum width in cloud radar, The atmospheric temperature is measured in the microwave radiometer. The relative humidity in the microwave radiometer;

[0097] S32. Calculation of the mean of the sample set:

[0098] ;

[0099] S33. Calculation of standard deviation of sample set:

[0100] ;

[0101] S34. Determination of the discrimination threshold: Based on the observation characteristics of extreme cold snowfall, take... This serves as the basic interval for the corresponding phase state parameters, i.e., the initial threshold.

[0102] S35. Repeat steps 31 to S34 to determine the initial thresholds for each parameter, and complete the discrimination of the threshold intervals through independent observation sample verification and physical rationality verification. The threshold interval discrimination conditions are as follows:

[0103] For ice crystals, the threshold range is as follows: ;

[0104] For dry snow, the threshold range is as follows:

[0105] For graupel, the threshold range is as follows: ;

[0106] For mixed phases, the threshold range is as follows: ;

[0107] For wet snow, the threshold range is as follows: .

[0108] S4. Inverting the instantaneous snowfall rate for a single time: First, calculate the instantaneous snowfall intensity at different altitude levels, then obtain the overall snowfall rate by vertical weighting and fusion. The specific operation for inverting the instantaneous snowfall rate for a single time is as follows:

[0109] S41. Traverse each distance database along the cloud radar profile and calculate the instantaneous snowfall intensity at different altitudes. During traversal, first check the particle phase classification results of the current distance database; only calculate the snowfall rate when the phase is wet snow or dry snow. The calculation method adopts... Power-law relationship The relationship is transformed into:

[0110] ;

[0111] In the formula, The current distance corresponds to the cloud radar reflectivity. This refers to the intensity of a single layer of instantaneous snowfall. , These are empirical coefficients corresponding to different phase states, typically for dry snow. wet snow It can be localized according to the geographical features and snowfall characteristics of different extremely cold regions;

[0112] S42. The instantaneous snowfall intensity of a single layer is the hydrophobicity formation rate of snow particles within that layer, while the ground snowfall rate is the fusion result of snow particles from different altitude layers after falling to the ground. Therefore, a vertical weighted fusion of instantaneous snowfall from each layer is necessary. High-altitude snow particles are prone to sublimation and evaporation during their descent, making it difficult for them to reach the ground; low-altitude snow particles are more physically stable and more likely to reach the ground. Therefore, different weights are assigned to different altitude layers, with a higher weight for lower altitudes.

[0113] ;

[0114] In the formula, For the entire layer of snowfall rate, For the first The instantaneous snowfall intensity at a distance from the reservoir. For the first The height contribution weight of each distance library For the first The height of the distance to the warehouse, The maximum height of the effective snow layer. Ground height; the effective snow layer refers to the vertical height interval defined by the distances of the first and last effectively identified as dry or wet snow within the current vertical profile.

[0115] S5. Cumulative Total Snowfall Over a Period: The cumulative snowfall is obtained by integrating the total snowfall rate obtained in step S4 over time. The specific method for calculating the cumulative total snowfall over a period is as follows:

[0116] If the cloud radar observation time series is The corresponding total snowfall rate Then the time period Cumulative snowfall Cumulative snowfall The calculation formula is:

[0117] .

[0118] When there is no data available, the cumulative snowfall is calculated directly by time integration; if there is data missing, the cumulative snowfall is calculated using discrete integration. The calculation formula is:

[0119] ;

[0120] In the formula, For the first Each observation time, For the first Each observation time, , for the first The length of each time period; for Snowfall rate per period for Snowfall rate per period The total number of valid observations within the target time period; This represents the cumulative snowfall during the specified period.

[0121] S6. Output the results.

[0122] This embodiment utilizes millimeter-wave cloud radar and microwave radiometer to conduct joint inversion of multi-source data. Addressing the spatiotemporal heterogeneity issue, it establishes spatiotemporal dimension matching rules based on cloud radar, achieving precise spatiotemporal alignment of millimeter-wave cloud radar and microwave radiometer observation data. Then, based on statistical analysis of multiple parameters such as reflectivity, radial velocity, temperature, and humidity, it constructs a particle phase classification system for extremely cold environments, accurately distinguishing five types of snowfall-related particles: ice crystals, dry snow, graupel, mixed phase, and wet snow, effectively eliminating interference from non-snowfall particles. Finally, it employs phase-specific methods for dry and wet snow. The power-law relationship is used to invert the snowfall rate, and the calculation of the snowfall rate of the entire layer is achieved through vertical weighted fusion.

Claims

1. A method for inverting snowfall based on millimeter-wave cloud radar and microwave radiometer, characterized in that, Includes the following steps: S1. Data Input: Input millimeter-wave cloud radar data and microwave radiometer data into the data processing module; S2. Multi-source data spatiotemporal dimension matching: The data processing module sequentially matches and aligns millimeter-wave cloud radar data with microwave radiometer data from both time and space dimensions. S3. Extreme cold particle phase classification: By collecting observation samples of continuous snowfall processes in extremely cold regions, a parameter sample set of various extreme cold particle phases is constructed. The initial discrimination threshold of each parameter is determined based on statistical analysis, and the final threshold is determined through actual measurement, calibration and verification to achieve extreme cold particle phase classification. S4. Inverting snowfall rates at different altitude levels: First, calculate the instantaneous snowfall intensity at different altitude levels for each layer, then obtain the overall snowfall rate for the entire layer through vertical weighted fusion. The specific operation is as follows: S41. Traverse each distance database along the cloud radar profile and calculate the instantaneous snowfall intensity at different altitudes. During traversal, first check the particle phase classification results of the current distance database; only calculate the snowfall rate when the phase is wet snow or dry snow. The calculation method adopts... Power-law relationship The relationship is transformed into: ； In the formula, The current distance corresponds to the cloud radar reflectivity. This refers to the intensity of a single layer of instantaneous snowfall. , These are the empirical coefficients corresponding to different phase states; S42. The instantaneous snowfall intensity of a single layer is the hydrophobicity formation rate of snow particles within that layer, while the ground snowfall rate is the fusion result of snow particles from different altitude layers after falling to the ground. Therefore, a vertical weighted fusion of instantaneous snowfall from each layer is necessary. High-altitude snow particles are prone to sublimation and evaporation during their descent, making it difficult for them to reach the ground; low-altitude snow particles are more physically stable and more likely to reach the ground. Therefore, different weights are assigned to different altitude layers, with a higher weight for lower altitudes. ； In the formula, For the entire layer of snowfall rate, For the first The instantaneous snowfall intensity at a distance from the reservoir. For the first The height contribution weight of each distance library For the first The height of the distance to the warehouse, The maximum height of the effective snow layer. Ground height; the effective snow layer refers to the vertical height interval defined by the distances of the first and last effectively identified as dry or wet snow in the current vertical profile; S5. Calculate the total cumulative snowfall over the period: The cumulative snowfall is obtained by integrating the whole-layer snowfall rate obtained in step S4 over time. S6. Output the results.

2. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 1, characterized in that, In step S2, the time resolutions of cloud radar data and microwave radiometer data are different in the time dimension; the time resolution of cloud radar data is selected as the time reference, and the specific operation is as follows: For each cloud radar time Microwave radiation time with the closest matching time , ,like If the two sets of data match successfully, then the two sets of data are considered to be a match.

3. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 1, characterized in that, In step S2, the spatial differences between cloud radar data and microwave radiometer data are more significant. Since the spatial resolution of cloud radar data is higher than that of microwave radiometer data, the specific operation is based on the spatial resolution of cloud radar data as follows: The altitude corresponding to each distance in the cloud radar database is used as a sub-element, and the data are sorted in ascending order to generate a baseline altitude profile. , ； In the formula, As the reference height profile, ,..., These represent the heights corresponding to each distance in the cloud radar database, such as: Indicates the first The height corresponding to each library; The microwave radiometer data was then interpolated to the reference height profile. Above, for any microwave radiometer variable At the reference height The interpolation result at point is: ； In the formula, , Distance in microwave radiometer data The two most recent effective observation altitudes meet the requirements. ; For height Microwave radiometer observations For height Microwave radiometer observations For height Interpolation results of the microwave radiometer.

4. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 1, characterized in that, In step S3, in order to eliminate the error caused by non-snowfall particles in the snowfall rate inversion, particle phase classification was performed before the snowfall amount inversion. Since the application scenario of snowfall amount inversion is extreme cold snowfall weather, the extreme cold particle phase classification is only carried out for the relevant phases in extreme cold environments. By collecting observation samples of continuous snowfall processes in extreme cold regions, a parameter sample set of various extreme cold particle phases is constructed. The initial discrimination threshold of each parameter is determined based on statistical analysis, and the final threshold is determined by actual measurement calibration and verification.

5. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 4, characterized in that, Based on the inherent differences in physical structure, motion characteristics, and environmental conditions of different particles, the reflectivity in wave cloud radar is used as an example. radial velocity Velocity spectral width and atmospheric temperature in microwave radiometers and relative humidity As a basis for judgment, the system uses multi-condition combined logic to distinguish the five extreme cold particle phases: ice crystals, dry snow, graupel, mixed phase, and wet snow.

6. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 5, characterized in that, The specific steps for distinguishing the phase states of the five types of extremely cold particles are as follows: S31. Definition of parameter sample set: For the first... Phase-like states, These correspond to ice crystals, dry snow, graupel, mixed phase, and wet snow, respectively, and are collected. Construct a parameter sample set from 100 valid observation samples: In the formula, Reflectivity in cloud radar For radial velocity in cloud radar, For the velocity spectrum width in cloud radar, The atmospheric temperature is measured in the microwave radiometer. The relative humidity in the microwave radiometer; S32. Calculation of the mean of the sample set: ； S33. Calculation of standard deviation of sample set: ； S34. Determination of the discrimination threshold: Based on the observation characteristics of extreme cold snowfall, take... This serves as the basic interval for the corresponding phase state parameters, i.e., the initial threshold. S35. Repeat steps S31 to S34 to determine the initial threshold of each parameter, and complete the determination of the threshold range through independent observation sample verification and physical rationality verification.

7. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 6, characterized in that, In step S35, for the initial threshold of each parameter, the threshold interval discrimination conditions are determined through independent observation sample verification and physical rationality verification as follows: For ice crystals, the threshold range is as follows: ; For dry snow, the threshold range is as follows: ; For graupel, the threshold range is as follows: ; For mixed phases, the threshold range is as follows: ; For wet snow, the threshold range is as follows: .

8. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 7, characterized in that, In step S5, the specific method for accumulating the total snowfall over the time period is as follows: If the cloud radar observation time series is The corresponding total snowfall rate Then the time period Cumulative snowfall Cumulative snowfall When there is no data available, the calculation is performed by direct integration over time. The formula is as follows: 。 9. The snowfall inversion method based on millimeter-wave cloud radar and microwave radiometer according to claim 8, characterized in that, If there are missing data points, the cumulative snowfall will be calculated using discrete integral form. The calculation formula is: ； In the formula, For the first Each observation time, For the first Each observation time, , for the first The length of each time period; for Snowfall rate per period for Snowfall rate per period The total number of valid observations within the target time period; This represents the cumulative snowfall during the specified period.

Images