Three-dimensional live analysis field construction method, system and storage medium

Abstract

The present invention discloses a method, system and storage medium for constructing a three-dimensional real-time analysis field, which relates to the technical field of meteorological analysis. The method includes: obtaining three-dimensional forecast field data of a target area, and obtaining background field data with a target horizontal resolution after downscaling the three-dimensional forecast field; receiving multi-source observation data, which includes conventional observation data and remote sensing data; fusing the multi-source observation data into the high-resolution background field data to correct the background field and perform cloud analysis, so as to obtain a high-resolution three-dimensional real-time analysis field; wherein, the conventional meteorological element observations are assimilated into the background field by using the successive correction method to correct the model background field, and the complex cloud analysis method is used in combination with radar reflectivity observations to perform water substance phase analysis and adjust the temperature and humidity fields inside the clouds. The real-time analysis field constructed by the present invention has high resolution and high accuracy, and improves the accuracy of real-time analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological analysis, and in particular, to a method, system, and storage medium for constructing a three-dimensional real-time analysis field. Background Art

[0002] In megacities, local convection develops rapidly, and severe convective weather is prone to occur, leading to disasters such as hail, heavy rain, lightning, and strong winds. The accurate prediction of severe convective weather is a difficult problem in current meteorological research. To achieve refined prediction of severe convective weather, accurate real-time analysis is usually required first. Accurately describing the real-time state of the atmospheric system through real-time analysis has important scientific and practical value.

[0003] The initial conditions of the atmospheric system are the basis of numerical weather prediction (NWP), which are estimated by combining existing atmospheric knowledge (usually in the form of numerical models) with observational data through data assimilation (DA) technology. Data assimilation (DA) is a method of fusing new observational data during the dynamic operation process based on considering the spatio-temporal distribution of data and making error estimates for models and observations. The ultimate goal is to form a dataset with spatial, temporal, and physical consistency. Data assimilation can improve the estimation accuracy of the model state and enhance the prediction ability. At the same time, with the development of meteorological observation systems, the density of the observation network has increased year by year. In three-dimensional space, the construction of wind profilers, lidars, microwave radiometers, and Fengyun-4 series satellites has effectively filled the observation gaps.

[0004] With the great progress of data assimilation (DA) methods and the growth of observational data, the accuracy of real-time analysis has also been greatly improved. However, existing real-time analysis products / systems mainly use the forecast fields of mainstream business models as background fields and utilize data fusion technology to produce real-time analysis fields, which have the following defects: low spatial resolution, insufficient three-dimensional space detection, and the accuracy of the analysis field is difficult to meet the real-time analysis requirements of some key areas of concern. Summary of the Invention

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a method, system, and storage medium for constructing a three-dimensional real-time analysis field. Based on measured data and background field data, the present invention constructs a three-dimensional real-time analysis field with high resolution and high accuracy. First, the forecast field is downscaled to obtain a high-resolution background field, and then the stepwise correction and complex cloud analysis assimilation algorithm are used to fuse multi-source observational data for data assimilation analysis, improving the accuracy of the real-time analysis field; at the same time, the three-dimensional real-time analysis field can analyze the three-dimensional states of multiple variables.

[0006] To achieve the above objectives, the present invention provides the following technical solutions:

[0007] A method for constructing a three-dimensional real-time analysis field, the method comprising the following steps:

[0008] Obtain three-dimensional forecast field data of a target area, the horizontal resolution of the three-dimensional forecast field being the initial horizontal resolution;

[0009] After performing downscaling processing on the foregoing three-dimensional forecast field, obtain background field data with a target horizontal resolution; the value of the target horizontal resolution is less than the value of the initial horizontal resolution, such that the resolution of the background field is higher than that of the forecast field;

[0010] Receive multi-source observation data, the multi-source observation data including conventional observation data and remote sensing data;

[0011] Fuse the foregoing multi-source observation data into the high-resolution background field data to perform correction processing and cloud analysis on the background field, and obtain a high-resolution three-dimensional real-time analysis field; wherein, the conventional meteorological element observations are assimilated into the background field by the successive correction method to correct the model background field, and the complex cloud analysis method is used in combination with radar reflectivity observations to perform water substance phase analysis and adjustment of the temperature and humidity fields inside the clouds.

[0012] Furthermore, the conventional observation data includes real-time observation data of surface automatic stations, aircraft reports, and sounding stations;

[0013] The remote sensing data includes real-time observation data of radar reflectivity, wind profiler radar, FY4A retrieved temperature profile, and ground-based microwave radiometer temperature profile.

[0014] Furthermore, the conventional meteorological element observations are assimilated into the background field by the Bratseth successive correction method. For the nth iteration, the corresponding iteration expression is as follows:

[0015]

[0016] Wherein,

[0017] n is the number of iterations;

[0018] f i n is the value of the nth iteration at grid point i;

[0019] is the value of the (n - 1)th iteration at grid point i;

[0020] is the observed value of the kth observation point around grid point i;

[0021] nobs is the total number of grid points;

[0022] is the estimated value of the (n - 1)th iteration at observation point k, and this estimated value is obtained by grid interpolation;

[0023] α is a weight coefficient, which is determined by the variance of the observation error, the variance of the background error, and the distance from the grid point to the observation; α ik is the weight coefficient of the observation value of the k-th observation point around the grid point i.

[0024] Furthermore, after each iteration ends, balance adjustment is performed based on the preset balance constraint conditions;

[0025] And, according to the preset number of iteration cycles, when the number of iterations reaches the aforementioned number of iteration cycles, the update information of the actual observation data is obtained, and new actual observation data is introduced for another iteration.

[0026] Furthermore, the steps of the water substance phase analysis and the adjustment of the temperature and humidity fields inside the cloud are as follows:

[0027] Three-dimensional grid cloud amount calculation: According to the relative humidity information of the model background field, the three-dimensional background field cloud amount is calculated using a preset empirical relationship, and the three-dimensional background field cloud amount distribution is determined; then, the cloud base, cloud amount, and cloud thickness are corrected using the cloud observation data in the ground actual observation data; subsequently, the cloud amount is further adjusted using the model-gridded radar reflectivity observation data;

[0028] Construction and calculation of the three-dimensional cloud field: A three-dimensional cloud field is constructed based on the cloud base, cloud height, and grid cloud amount obtained in the previous step; after the construction of the three-dimensional cloud field is completed, the mass concentration of cloud droplet particles is estimated first to obtain the cloud water and cloud ice content at the grid point, and the cloud droplet particles are the sum of cloud ice and cloud water; and, in combination with the wet-bulb temperature and radar echo, the precipitation type is analyzed, and the analyzed precipitation types include rain, snow, freezing rain, ice rain, and hail; after the precipitation type is determined, according to the radar reflectivity observation data, the precipitation particle content of the model is quantitatively calculated to obtain the three-dimensional distribution of precipitation particles;

[0029] Adjustment of the model temperature and humidity fields: Based on the moist adiabatic or non-adiabatic initialization method, the model temperature and humidity fields are adjusted.

[0030] Furthermore, the specific steps of the three-dimensional grid cloud amount calculation are as follows:

[0031] Obtain the relative humidity information of the model background field, and calculate the cloud amount using the following empirical relationship according to the relative humidity information of the model background field to obtain the three-dimensional background field cloud amount CF,

[0032]

[0033] where CF is the cloud amount, and its value ranges from 0 to 1.0; RH is the relative humidity of the model background field; RH0 is a relative humidity threshold that varies with height; b is a preset empirical constant;

[0034] Then, using the cloud observation data of ground truth observation data, the cloud base, cloud amount and cloud thickness are corrected by the Barnes interpolation method; if there is no ground observation data related to clouds, no correction is made;

[0035] Combined with the cloud amount of the three-dimensional background field and the temperature of the initial field, a cloud top brightness temperature is calculated;

[0036] The cloud amount value is further adjusted according to the radar reflectivity observation data; among them, when the radar reflectivity of the grid point is greater than the set threshold and the radar echo is above the cloud base, the cloud amount value of the grid point is set to 1.

[0037] Furthermore, when constructing and calculating the three-dimensional cloud field, a cloud water and cloud ice calculation unit, a precipitation type calculation unit and a precipitation particle quantification calculation unit are configured;

[0038] The cloud water and cloud ice calculation is configured as follows: Obtain the grid cloud amount of each grid point in the three-dimensional cloud field, compare the grid cloud amount of each grid point with the preset cloud amount threshold, and for the grid points with grid cloud amount greater than the preset cloud amount threshold, calculate the cloud ice and cloud water content on these grid points. The steps are as follows: Divide the cloud into multiple layers from the cloud base to the cloud top, and calculate the temperature, air pressure and saturated water vapor pressure of each layer layer by layer to obtain the saturated water vapor mixing ratio. The temperature of each layer inside the cloud required in the calculation is calculated according to the saturated non-adiabatic vertical lapse rate, the air pressure of each layer is calculated according to the hypsometric formula, and the saturated water vapor pressure of each layer is calculated through the Clausius-Clapeyron equation; then calculate the difference in saturated water vapor mixing ratio between adjacent layers as the basic increment or decrement of cloud water and cloud ice. At this time, the cloud droplet particle mixing ratio ALWC of each layer is expressed as follows:

[0039] ALWC k =ALWC k-1 +q vs_k-1 -q vs_k

[0040] where k is the number of layers of cloud stratification inside the cloud; ALWC k is the cloud droplet particle mixing ratio of the kth layer; ALWC k-1 is the cloud droplet particle mixing ratio of the k-1th layer; q vs is the saturated water vapor mixing ratio, q vs_k is the saturated water vapor mixing ratio of the kth layer, q vs_k-1is the saturated water vapor mixing ratio of the (k - 1)th layer; after obtaining the cloud droplet particle mixing ratios of each layer, the cloud ice and cloud water contents are separated according to the ambient temperature T. Among them, when the temperature is above the first preset temperature threshold, it is determined to be all cloud water; when the temperature is below the second preset temperature threshold, it is determined to be all cloud ice; when the temperature is between the first preset temperature threshold and the second preset temperature threshold, it is determined to be a mixture of cloud ice and cloud water. Among them, the weight of cloud water is 0.05 * (T - the second preset temperature threshold), and the remaining part is cloud ice, where T represents the ambient temperature value;

[0041] The precipitation type calculation unit is configured to: starting from the echo top of each grid cell, when the wet-bulb temperature Tw of the echo top is lower than 0 °C, it is determined that the precipitate starts as snow, otherwise it is determined to be rain; when snow falls into an area where the wet-bulb temperature is higher than 1.3 °C, it is determined that the precipitate melts into rain; when rain falls to a place where the wet-bulb temperature is lower than 0 °C, it is determined that the precipitate condenses into freezing rain; when freezing rain falls between the preset pressure layers P1 and P2, it is determined that the precipitate becomes sleet, and the pressures P1 and P2 satisfy the expression , where T represents the ambient temperature value, |dP| represents the absolute value of the differential dP, mb represents millibar, which is a pressure unit; when the radar echo intensity exceeds a certain set threshold, the precipitation type is determined to be hail;

[0042] The precipitation particle quantitative calculation unit is configured to: obtain radar reflectivity observation data; based on the radar reflectivity observation data, quantitatively calculate the precipitation particle content based on the empirical relationship between radar reflectivity and different types of precipitation particles; among them, the total radar reflectivity includes the contributions of rain, snow, and hail to the reflectivity factor of different precipitation particles.

[0043] Furthermore, there are two calculation modes configured in the precipitation particle quantitative calculation, including the first calculation mode and the second calculation mode;

[0044] The first calculation mode is the Kessler mode based on the parameterization method of fitting observations. At this time, the calculation formula for the reflectivity factor of rain is:

[0045] Z = a(ρ·q r ) b

[0046] where Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10log 18 Z; a and b are preset empirical constants; ρ is the air density, with the unit of kg / m 3 ; q r is the rain mixing ratio, with the unit of g / kg;

[0047] The calculation formula for the reflectivity factor of snow and hail is as follows:

[0048] Z = c(ρ·q s ) d

[0049] where Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10log 18 Z; c and d are preset empirical constants; ρ is the air density, with the unit of kg / m 3 ; q s is the rain mixing ratio, with the unit of g / kg;

[0050] The second calculation mode is the Ferrier mode based on the cloud physical process and the backscattering method of hydrometeors. At this time, the relationship between the reflectivity factor equation and the model variables is:

[0051] Z e = Z er + Z es + Z eh

[0052] where Z e is the total reflectivity factor, with the unit of mm 6 / m 3 , including the contributions of the rain reflectivity factor Z er , the snow water reflectivity factor Z es and the hail reflectivity factor Z eh ;

[0053] For rain, the equation expression of its reflectivity factor Z er is:

[0054]

[0055] where ρ r represents the rain density, and ρ represents the air density; N r represents the intercept that satisfies the Marshall-Palmer drop size distribution; q r represents the rain mixing ratio;

[0056] For snow, it is divided into dry snow and wet snow. When the temperature is lower than 0°C, it is defined as dry snow, and vice versa for wet snow;

[0057] The equation expression of the reflectivity factor Z es of dry snow is:

[0058]

[0059] where ρ srepresents the density of snow, ρ represents the air density, ρ i represents the density of ice, N s represents the intercept of snow, K i 2 and K r 2 represent the dielectric constants of ice and water respectively, q s represents the mixing ratio of dry snow;

[0060] The reflectivity factor Z of wet snow es The equation expression is:

[0061]

[0062] wherein, ρ s represents the density of snow, ρ represents the air density, N s represents the intercept of snow, q s represents the mixing ratio of wet snow;

[0063] For hail, the reflectivity factor Z eh The equation expression is:

[0064]

[0065] wherein, ρ h represents the density of hail, ρ represents the air density, N h represents the intercept of hail, q h represents the mixing ratio of hail.

[0066] The present invention also provides a three-dimensional real-time analysis system, and the system includes:

[0067] A forecast field downscaling module, configured to receive three-dimensional forecast field data of a target area, and the horizontal resolution of the three-dimensional forecast field is an initial horizontal resolution; perform downscaling processing on the foregoing three-dimensional forecast field to obtain background field data with a target horizontal resolution; the value of the target horizontal resolution is less than the value of the initial horizontal resolution, so that the resolution of the background field is higher than that of the forecast field;

[0068] An observation data collection and processing module, configured to collect and collate multi-source observation data, and the multi-source observation data includes conventional observation data and remote sensing data;

[0069] An assimilation analysis module, configured to receive the multi-source observation data transmitted by the observation data collection and processing module, and fuse the foregoing multi-source observation data into the high-resolution background field data to perform correction processing and cloud analysis on the background field, so as to obtain a high-resolution three-dimensional real-time analysis field; wherein, the conventional meteorological element observations are assimilated into the background field by using the successive correction method to correct the model background field, and the complex cloud analysis method is combined with the radar reflectivity observation to perform water substance phase analysis and in-cloud temperature and humidity field adjustment;

[0070] A display module for outputting the aforementioned three-dimensional real-time analysis field.

[0071] The present invention also provides a computer-readable storage medium, on which computer program instructions are stored, and the computer program instructions, when executed by a processor, implement the method described in any one of the foregoing.

[0072] Due to the adoption of the above technical solutions, compared with the prior art, for example, the present invention has the following advantages and positive effects:

[0073] Based on measured data and background field data, the present invention constructs a high-resolution and high-precision three-dimensional real-time analysis field. First, the forecast field is downscaled to obtain a high-resolution background field, and then a stepwise correction and complex cloud analysis assimilation algorithm is used to fuse multi-source observation data for data assimilation analysis, improving the accuracy of the real-time analysis field; at the same time, the three-dimensional real-time analysis field can analyze the multi-variable three-dimensional state. Description of the Drawings

[0074] Figure 1 It is a schematic flowchart of the method for constructing a three-dimensional real-time analysis field provided by an embodiment of the present invention.

[0075] Figure 2 It is a schematic diagram of the construction process of the 1km real-time analysis field provided by an embodiment of the present invention.

[0076] Figure 3 It is a comparison diagram of the vertical distribution of the temperature root mean square error provided by an embodiment of the present invention.

[0077] Figure 4 It is a comparison diagram of the vertical distribution of the relative humidity root mean square error provided by an embodiment of the present invention.

[0078] Figure 5 It is a comparison diagram of the vertical distribution of the zonal wind root mean square error provided by an embodiment of the present invention.

[0079] Figure 6 It is a comparison diagram of the vertical distribution of the meridional wind root mean square error provided by an embodiment of the present invention. Detailed Description of the Embodiments

[0080] The following further elaborates in detail on the method, system, and storage medium for constructing a three-dimensional real-time analysis field disclosed by the present invention in conjunction with the accompanying drawings and specific embodiments. It should be noted that the technical features described in the following embodiments or the combination of technical features should not be considered in isolation, and they can be combined with each other to achieve better technical effects. In the accompanying drawings of the following embodiments, the same reference numerals in each drawing represent the same features or components, which can be applied to different embodiments. Therefore, once a certain item is defined in one drawing, it does not need to be further discussed in the subsequent drawings.

[0081] It should be noted that the structures, proportions, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the conditions for the implementation of the invention. Any modification of the structure, change in the proportional relationship or adjustment of the size, without affecting the effects that the invention can produce and the purposes that can be achieved, should fall within the scope covered by the technical content disclosed by the invention. The scope of the preferred implementation of the present invention includes additional implementations, in which the functions can be executed in a substantially simultaneous manner or in the reverse order according to the functions involved, rather than in the order described or discussed, which should be understood by those skilled in the technical field to which the embodiments of the present invention belong.

[0082] Techniques, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be regarded as an integral part of the authorized specification. In all examples shown and discussed here, any specific values should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0083] Explanation of technical terms:

[0084] ERA5: It is the fifth-generation reanalysis product launched by the European Centre for Medium-Range Weather Forecasts. It integrates a variety of observational data and numerical model data, providing high-resolution climate variable data on a global scale. These data include climate variables of the atmosphere, land, and ocean, with a time resolution of up to hourly and a spatial resolution on a 30 km grid. ERA5 reanalysis data has a wide range of applications in multiple fields such as meteorology, climatology, and hydrology, and it is also one of the publicly available reanalysis datasets with relatively high quality at present.

[0085] Embodiment

[0086] See Figure 1 As shown, it is a method for constructing a three-dimensional real-time analysis field provided by this embodiment. The method includes the following steps.

[0087] S100, obtain the three-dimensional forecast field data of the target area. The horizontal resolution of the three-dimensional forecast field is the initial horizontal resolution.

[0088] S200, perform downscaling processing on the aforementioned three-dimensional forecast field to obtain background field data with a target horizontal resolution; the value of the target horizontal resolution is less than the value of the initial horizontal resolution, so that the resolution of the background field is higher than that of the forecast field.

[0089] The target horizontal resolution can be set by the system default or by the user according to the observational needs of the key area of concern.

[0090] S300, receive multi-source observation data.

[0091] In this embodiment, the multi-source observation data includes conventional observation data and remote sensing data. Preferably, the conventional observation data may include the actual observation data of surface automatic stations, aircraft reports, and radiosonde stations. The remote sensing data may include the actual observation data of radar reflectivity, wind profiler radar, FY4A retrieved temperature profile, and ground-based microwave radiometer temperature profile.

[0092] S400, fuse the foregoing multi-source observation data into high-resolution background field data to perform correction processing and cloud analysis on the background field, and obtain a high-resolution three-dimensional actual analysis field. Among them, the successive correction method is used to assimilate the conventional meteorological element observations (such as wind, pressure, temperature, humidity, etc.) into the background field to correct the model background field, and a complex cloud analysis method is used in combination with radar reflectivity observations to perform water substance phase analysis and adjustment of the temperature and humidity fields inside the cloud.

[0093] In this embodiment, the successive correction method in step S400 may specifically adopt the Bratseth method. The Bratseth method does not require a huge calculation matrix, and the iterative method can save a great deal of calculation time to a large extent.

[0094] Assimilate the conventional meteorological element observations (wind, pressure, temperature, humidity) into the background field through the Bratseth successive correction method. In this way, the information of the conventional meteorological element observations is used to correct the model background field to make it more accurate. There are multiple analysis variables in the background field. Among them, the main analysis variables include u and v horizontal wind components, air pressure, potential temperature, specific humidity, etc.

[0095] At this time, for the nth iteration, the corresponding iteration expression is as follows:

[0096]

[0097] where n is the number of iterations; f i n is the nth iteration value at grid point i; is the (n - 1)th iteration value at grid point i; is the observed value of the kth observation point around grid point i; nobs is the total number of grid points; is the (n - 1)th estimated value at observation point k, and this estimated value is obtained through grid interpolation; α is the weight coefficient, which is determined by the observation error variance, background error variance, and the distance from the grid point to the observation; α ik is the weight coefficient of the observed value of the kth observation point around grid point i.

[0098] After each iteration, adjustments such as balance can be made, and after several iterations, new and more observational data can be introduced for another iteration. Specifically, balance constraint conditions and the number of iteration cycles can be preset; after each iteration, balance adjustment is performed based on the preset balance constraint conditions; and when the number of iterations reaches the preset number of iteration cycles, update information of the actual observational data is obtained, and new actual observational data is introduced for another iteration. The number of iteration cycles is used to set the number of iterations in one round of iteration. After new observational data is introduced, a new round of iteration begins. That is to say, when new observational data is received, the previous round of iteration ends, and a new round of iteration begins based on the introduced new observational data.

[0099] For the cloud analysis method, in this embodiment, a complex cloud analysis method is adopted for water substance phase analysis and adjustment of the temperature and humidity fields inside the cloud. First, using the model-gridded radar reflectivity, the background cloud amount is calculated according to the relative humidity information of the model background field by using an empirical relationship, so as to determine the three-dimensional background cloud amount distribution and correct the cloud base, cloud amount, and cloud thickness; then, the precipitation type is analyzed by combining the wet-bulb temperature and radar echo, and the analyzed precipitation types include rain, snow, freezing rain, ice rain, and hail; at the same time, based on the radar reflectivity observation, the three-dimensional distribution of the precipitation particles of the model is quantitatively calculated, so as to construct and adjust the mass mixing ratio of the initial three-dimensional cloud and precipitation particles of the model. Subsequently, based on the wet adiabatic or non-adiabatic initialization method, the temperature and humidity fields of the model are further adjusted.

[0100] Specifically, the steps of water substance phase analysis and adjustment of the temperature and humidity fields inside the cloud in step S400 can be as follows:

[0101] S410, Three-dimensional grid cloud amount calculation: According to the relative humidity information of the model background field, the three-dimensional background cloud amount is calculated by using a preset empirical relationship to determine the three-dimensional background cloud amount distribution; then, the cloud base, cloud amount, and cloud thickness are corrected by using the cloud observation data in the ground actual observational data; subsequently, the cloud amount is further adjusted by using the model-gridded radar reflectivity observational data.

[0102] S420, Construction and calculation of the three-dimensional cloud field: A three-dimensional cloud field is constructed according to the cloud base, cloud height, and grid cloud amount obtained in the previous step; after the construction of the three-dimensional cloud field is completed, the mass concentration of cloud droplet particles is estimated first to obtain the cloud water and cloud ice content at the grid points, and the cloud droplet particles are the sum of cloud ice and cloud water; and the precipitation type is analyzed by combining the wet-bulb temperature and radar echo, and the analyzed precipitation types include rain, snow, freezing rain, ice rain, and hail; after the precipitation type is determined, according to the radar reflectivity observational data, the precipitation particle content of the model is quantitatively calculated to obtain the three-dimensional distribution of the precipitation particles.

[0103] S430, Adjustment of the model temperature and humidity fields: Based on the moist adiabatic or non-adiabatic initialization method, the model temperature and humidity fields are adjusted. Among them, when adjusting the temperature inside the cloud, a latent heat release scheme, a temperature adjustment scheme based on the moist adiabatic profile, or a buoyancy potential temperature adjustment scheme is adopted; when adjusting the relative humidity and water vapor inside the cloud, according to the calculation result of the three-dimensional grid cloud amount, the relative humidity is first adjusted, and then the water vapor field is further adjusted according to the adjustment result of the relative humidity.

[0104] When constructing and calculating the three-dimensional cloud field in step S420, a cloud water and cloud ice calculation unit, a precipitation type calculation unit, and a precipitation particle quantification calculation unit can be respectively configured.

[0105] The cloud water and cloud ice calculation is configured to: obtain the grid cloud amount of each grid point in the three-dimensional cloud field, compare the grid cloud amount of each grid point with a preset cloud amount threshold, and for the grid points where the grid cloud amount is greater than the preset cloud amount threshold, calculate the cloud ice and cloud water content on these grid points.

[0106] The precipitation type calculation unit is configured to: analyze the precipitation type by combining the wet bulb temperature and radar echo, and the analyzed precipitation types include rain, snow, freezing rain, ice rain, and hail.

[0107] The precipitation particle quantification calculation unit is configured to: obtain radar reflectivity observation data; based on the radar reflectivity observation data and the empirical relationship between radar reflectivity and different types of precipitation particles, quantitatively calculate the precipitation particle content; where the total radar reflectivity includes the contributions of different precipitation particles such as rain, snow, and hail to the reflectivity factor.

[0108] Preferably, multiple calculation modes can be configured in the precipitation particle quantification calculation unit in this embodiment, and the user can select multiple calculation modes as needed during use to perform multi-scheme simulations of the precipitation situation.

[0109] The following combines Figure 2 , taking the construction of a 1 km real-time analysis field in a certain target area as an example, to describe in detail the complex cloud analysis process of this embodiment.

[0110] Taking a key area of concern in the East China region as the target area, first obtain a three-dimensional forecast field with a horizontal resolution of 3 km for the target area through the CMA-SH3 rapid update and assimilation system in the East China region. Then, perform downscaling on the aforementioned three-dimensional forecast field through the forecast field downscaling module to obtain a background field with a horizontal resolution of 1 km. Then, use the assimilation analysis module to fuse the received multi-source observation data (which can include the actual observation data of surface automatic stations, radar reflectivity, wind profiler radars, FY4A retrieved temperature profiles, ground-based microwave radiometer temperature profiles, aircraft reports, sounding stations, etc.) into the 1 km background field data to correct the background field and perform cloud analysis, thereby constructing a 1 km three-dimensional actual analysis field for the target area, and the time resolution of this actual analysis field can be configured to 1 hour.

[0111] First, calculate the three-dimensional grid cloud amount.

[0112] The specific steps are as follows:

[0113] S411, obtain the relative humidity information of the model background field, and calculate the cloud amount according to the following empirical relationship based on the relative humidity information of the model background field to obtain the three-dimensional background field cloud amount CF,

[0114]

[0115] where CF is the cloud amount, and its value ranges from 0 to 1.0; RH is the relative humidity of the model background field; RH0 is a relative humidity threshold that varies with height; b is a preset empirical constant, which takes the value of 2 in this embodiment.

[0116] S412, use the cloud observation data of the ground actual observation data - including cloud base, cloud height, etc., to correct the cloud base, cloud amount and cloud thickness through the Barnes interpolation method; if there is no ground observation data related to clouds, no correction is performed.

[0117] S413, according to the three-dimensional background field cloud amount obtained previously, combined with the temperature of the initial field, a cloud top bright temperature (i.e., the temperature at the top of the cloud) can be deduced.

[0118] S414, further adjust the cloud amount value according to the radar reflectivity observation data. Among them, when the radar reflectivity of the grid point is greater than the set threshold and the radar echo is above the cloud base, the cloud amount value of this grid point is set to 1.

[0119] Secondly, perform cloud water and cloud ice calculations.

[0120] After completing the construction of the three-dimensional cloud field by combining the background field and the observation data (that is, determining the cloud base, cloud height and grid cloud amount), estimate the mass concentration of cloud droplet particles to obtain the cloud water and cloud ice content at the grid point, and the cloud droplet particles are the sum of cloud ice and cloud water.

[0121] Specifically, for grid points where the grid cloud amount is greater than a preset cloud amount threshold, such as 0.65, the cloud ice and cloud water content are calculated according to the Smith-Feddes scheme at these grid points. The specific steps are as follows:

[0122] The cloud is divided into multiple layers from the cloud base to the cloud top. For example, it is divided into layers every 100 m. The saturated water vapor mixing ratio is obtained layer by layer by calculating the temperature, air pressure, and saturated water vapor pressure. Then, the difference in the saturated water vapor mixing ratio between adjacent layers is calculated as the basic increment or decrement of cloud water and cloud ice. At this time, the cloud droplet particle mixing ratio ALWC of each layer is expressed as follows:

[0123] ALWC k = ALWC k-1 + q vs_k-1 - q vs_k

[0124] where k is the number of layers divided within the cloud; ALWC k is the cloud droplet particle mixing ratio of the kth layer; ALWC k-1 is the cloud droplet particle mixing ratio of the (k - 1)th layer; q vs is the saturated water vapor mixing ratio, q vs_k is the saturated water vapor mixing ratio of the kth layer, q vs_k-1 is the saturated water vapor mixing ratio of the (k - 1)th layer.

[0125] The temperature of each layer within the cloud required for the calculation is calculated according to the saturated non-adiabatic vertical lapse rate, the air pressure of each layer is calculated according to the hypsometric formula, and the saturated water vapor pressure is calculated through the Clausius-Clapeyron equation.

[0126] After obtaining the mixing ratio of cloud droplets (the sum of cloud ice and cloud water), the cloud ice and cloud water content are separated according to the ambient temperature T. Specifically, when the temperature is above 268.15 K, it can be determined that it is all cloud water; when the temperature is below 248.15 K, it can be determined that it is all cloud ice; and when the temperature is between 268.15 - 248.15 K, it is determined to be a mixture of cloud ice and cloud water. At this time, the weight of cloud water can be calculated by the formula 0.05*(T - 248.15), and the remaining part is cloud ice, representing the ambient temperature value.

[0127] Then, the precipitation type is calculated.

[0128] Analyze the three-dimensional precipitation types by combining the wet-bulb temperature and radar echo. The precipitation types to be analyzed include rain, snow, freezing rain, ice rain, and hail. The identification (diagnosis) of precipitation types starts from the echo top of each grid cell. If the wet-bulb temperature (Tw) at the echo top is lower than 0°C, it is determined that the precipitate is snow at the beginning, otherwise it is determined to be rain. When snow falls into an area where the wet-bulb temperature is higher than 1.3°C, it is determined that the precipitate melts into rain; when rain falls into a place where the wet-bulb temperature is lower than 0°C, it is determined that the precipitate condenses into freezing rain; when freezing rain falls between the preset pressure layers P1 and P2, it is determined that the precipitate becomes sleet, and the pressures P1 and P2 satisfy the expression , where T represents the ambient temperature value, |dP| represents the absolute value of the differential dP, and mb represents millibar, which is a pressure unit; when the radar echo intensity exceeds a certain set threshold, the precipitation type is determined to be hail.

[0129] Subsequently, quantitative calculation of precipitation particles is carried out.

[0130] In this embodiment, the quantitative calculation of precipitation particle content is mainly carried out according to the radar reflectivity observation data and the empirical relationships of different types of precipitation particles, and the radar reflectivity and precipitation-related factors such as precipitation particles are related through the reflectivity factor equation. The total radar reflectivity can specifically include the contributions of different precipitation particles such as rainwater, snow water, and hail to the reflectivity factor.

[0131] Preferably, in the precipitation particle quantitative calculation unit of this embodiment, two calculation modes can be configured, including the first calculation mode and the second calculation mode. The first calculation mode is the Kessler mode based on the parameterization method of fitting observations, and the second calculation mode is the Ferrier mode based on the cloud physical process and the backscattering method of hydrometeors.

[0132] Under the Kessler mode, the Kessler precipitation particle calculation scheme is adopted. At this time, the calculation formula for the reflectivity factor of rainwater is:

[0133] Z = a(ρ·q r ) b

[0134] where Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10log 18 Z; a = 17300.0, b = 7 / 4; ρ is the air density, with the unit of kg / m 3 ; q r is the rain mixing ratio, with the unit of g / kg.

[0135] The calculation formulas for the reflectivity factors of snow and hail are:

[0136] Z = c(ρ·q s ) d

[0137] where Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10log 18 Z; c = 38000.0, d = 2.2; ρ is the air density, with the unit of kg / m 3 ; q s is the rain water mixing ratio, with the unit of g / kg.

[0138] In the Ferrier model, the Ferrier precipitation particle calculation scheme is adopted. At this time, the relationship between the reflectivity factor equation and the model variables is:

[0139] Z e = Z er + Z es + Z eh

[0140] where Z e is the total reflectivity factor, with the unit of mm 6 / m 3 , including the contribution of the rain water reflectivity factor Z er , the snow water reflectivity factor Z es and the hail reflectivity factor Z eh .

[0141] For rain water, the equation expression of its reflectivity factor Z er is:

[0142]

[0143] where ρ r represents the rain water density, taking ρ r = 1000 kg / m 3 ; ρ represents the air density, with the unit of kg / m 3 ; N r represents the intercept satisfying the Marshall-Palmer drop size distribution, which is a fixed value in this embodiment, 8.0×10 6 m -4 ; q r represents the rain water mixing ratio, with the unit of g / kg.

[0144] For snow, it is divided into dry snow and wet snow. It is defined that when the temperature is lower than 0°C, it is dry snow, otherwise it is wet snow.

[0145] The equation expression of the reflectivity factor Z es of dry snow is:

[0146]

[0147] Among them, ρ s represents the density of snow, and ρ is taken as s = 100 kg / m 3 ; ρ represents the air density, with the unit kg / m 3 ; ρ i represents the density of ice, and ρ is taken as i = 917 kg / m 3 ; N s represents the intercept of snow, and N is taken as s = 3.0×10 6 m -4 ; K i 2 and K r 2 respectively represent the dielectric constants of ice and water, which are fixed values in this embodiment, 0.176 and 0.93 respectively; q s represents the mixing ratio of dry snow.

[0148] The reflectivity factor Z of wet snow es The equation expression is:

[0149]

[0150] Among them, ρ s represents the density of snow, ρ represents the air density, N s represents the intercept of snow, and the values of the three parameters are the same as above; q s represents the mixing ratio of wet snow.

[0151] For hail, the reflectivity factor Z eh The equation expression is:

[0152]

[0153] Among them, ρ h represents the density of hail, and ρ is taken as h = 917 kg / m 3 ; ρ represents the air density, with the unit kg / m 3 ; N h represents the intercept of hail, and N is taken as h = 4.0×10 4 m -4 ; q h represents the mixing ratio of hail.

[0154] After the precipitation type is determined, the mixing ratios of precipitation particles such as rain, snow, and hail are calculated from the above reflectivity factor equations.

[0155] Finally, the model temperature and humidity fields are adjusted.

[0156] It mainly includes the non-adiabatic initialization of the in-cloud temperature and the readjustment of the in-cloud relative humidity and water vapor.

[0157] For the non-adiabatic initialization of the in-cloud temperature, this embodiment provides multiple methods, including the latent heat release scheme, the temperature adjustment scheme based on the wet adiabatic profile, and the buoyancy potential temperature adjustment scheme, for users to choose from.

[0158] For the adjustment of the in-cloud relative humidity and water vapor, it is mainly based on the analysis and calculation results of the three-dimensional grid cloud amount. First, the humidity field is adjusted, and then the water vapor field is further adjusted according to the adjustment results of the humidity field.

[0159] Optionally, in this embodiment, after the three-dimensional real-time analysis field of the target area is constructed, it may further include the step of evaluating the accuracy of the three-dimensional real-time analysis field product. Specifically, the accuracy evaluation step of the real-time analysis field product includes: obtaining the actual detection data (such as sounding station data) of the target area and using it as the true value, and then comparing and analyzing the three-dimensional real-time analysis field product and the existing reanalysis product (such as ERA5) with the aforementioned true value respectively to obtain the root mean square error (RMSE) of the two on different analysis variables, which is used as the accuracy evaluation result.

[0160] By way of example and not limitation, for example Figures 3 to 5 The comparison results of a certain target area in the East China region are exemplified, including the vertical distribution of the root mean square errors of temperature, relative humidity, zonal wind, and meridional wind. It can be seen that the error distribution of the three-dimensional real-time analysis field product of this target area constructed based on this embodiment has good consistency with that of the internationally advanced ERA5 reanalysis tool, and the temperature and humidity fields are generally better than ERA5.

[0161] Another embodiment of the present invention also provides a three-dimensional real-time analysis system.

[0162] The system includes a forecast field downscaling module, an observation data collection and processing module, an assimilation analysis module, a display module, and an optional numerical weather prediction tool.

[0163] The numerical weather prediction tool is used to generate three-dimensional forecast field data for the target area, and the horizontal resolution of the three-dimensional forecast field is the initial horizontal resolution.

[0164] The forecast field downscaling module is used to receive the three-dimensional forecast field data of the target area, and the horizontal resolution of the three-dimensional forecast field is the initial horizontal resolution; perform downscaling processing on the aforementioned three-dimensional forecast field to obtain background field data with a target horizontal resolution; the value of the target horizontal resolution is less than the value of the initial horizontal resolution, so that the resolution of the background field is higher than that of the forecast field.

[0165] The observation data collection and processing module is used to collect and collate multi-source observation data, and the multi-source observation data includes conventional observation data and remote sensing data.

[0166] The assimilation analysis module is used to receive the multi-source observation data transmitted by the observation data collection and processing module, and fuse the multi-source observation data into the high-resolution background field data to correct the background field and perform cloud analysis, so as to obtain a high-resolution three-dimensional real-time analysis field; among them, the stepwise correction method is used to assimilate the conventional meteorological element observations into the background field to correct the model background field, and a complex cloud analysis method is used to perform water substance phase analysis and in-cloud temperature and humidity field adjustment in combination with radar reflectivity observations.

[0167] The display module is used to output the aforementioned three-dimensional real-time analysis field.

[0168] The conventional observation data may specifically include the real-time observation data of ground automatic stations, aircraft reports, and sounding stations.

[0169] The remote sensing data may specifically include the real-time observation data of radar reflectivity, wind profiler radar, FY4A-inverted temperature profile, and ground-based microwave radiometer temperature profile.

[0170] The stepwise correction method may specifically adopt the Bratseth method. The Bratseth method does not require a huge calculation matrix, and the iterative method can save a large amount of calculation time to a great extent. By using the Bratseth stepwise correction method, the conventional meteorological element observations (wind pressure, temperature, and humidity) are assimilated into the background field. In this way, the information of the conventional meteorological element observations is used to correct the model background field to make it more accurate. There are multiple analysis variables in the background field. Among them, the main analysis variables may include u and v horizontal wind components, air pressure, potential temperature, specific humidity, etc.

[0171] The steps of the water substance phase analysis and in-cloud temperature and humidity field adjustment may specifically include the three-dimensional grid cloud amount calculation step, the construction and calculation step of the three-dimensional cloud field, and the adjustment step of the model temperature and humidity field.

[0172] For other technical features, please refer to the description of the previous embodiments and will not be elaborated here.

[0173] Another embodiment of the present invention also provides a computer-readable storage medium for storing computer program instructions that can be executed by a processor. When the computer program instructions are executed by the processor, the method described above is implemented.

[0174] The storage medium may include various media that can store program codes, such as USB flash drives, mobile hard disks, read only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0175] For other technical features, reference may be made to the description of the foregoing embodiments, which will not be repeated herein.

[0176] In the above description, the disclosure of the present invention is not intended to limit itself to these aspects. Instead, within the scope of the object of the present disclosure, the components may be selectively and operably combined in any number. Additionally, terms such as "including", "comprising", and "having" should be construed as inclusive or open by default, rather than exclusive or closed, unless it is explicitly defined to the contrary. All technical, technological, or other terms conform to the meanings understood by those skilled in the art, unless it is defined to the contrary. Common terms found in dictionaries should not be interpreted too idealistically or too unrealistically in the context of relevant technical documents, unless the present disclosure clearly defines them as such. Any changes or modifications made by those of ordinary skill in the art of the present invention based on the above disclosure fall within the scope of protection of the claims.

Claims

1. A method for constructing a three-dimensional live analysis field, characterized in that Including the steps of: Obtaining three-dimensional forecast field data of the target area, with an initial horizontal resolution; After downscaling the three-dimensional forecast field, background field data with a target horizontal resolution is obtained. The value of the target horizontal resolution is less than the initial horizontal resolution, making the resolution of the background field higher than that of the forecast field; Receiving multi-source observation data, including conventional observation data and remote sensing data; Fusing the multi-source observation data into the high-resolution background field data to correct the background field and perform cloud analysis, obtaining a high-resolution three-dimensional real-time analysis field; among them, the conventional meteorological element observations are assimilated into the background field using the successive correction method to correct the model background field, and Combined with radar reflectivity observations, a complex cloud analysis method is used for water substance phase analysis and adjustment of the temperature and humidity fields inside the cloud, including the steps of: calculating three-dimensional grid cloud amounts, constructing and calculating the three-dimensional cloud field, and adjusting the model temperature and humidity fields; the construction and calculation of the three-dimensional cloud field include: after constructing the three-dimensional cloud field based on the cloud base, cloud top, and grid cloud amounts obtained in the previous steps, estimating the mass concentration of cloud droplet particles to obtain the cloud water and cloud ice content at the grid points. The cloud droplet particles are the sum of cloud ice and cloud water. Among them, the grid cloud amounts of each grid point in the three-dimensional cloud field are obtained, and the grid cloud amounts of each grid point are compared with a preset cloud amount threshold. For grid points with grid cloud amounts greater than the preset cloud amount threshold, calculate the cloud ice content and cloud water content at these grid points; and, analyzing the precipitation type in combination with the wet-bulb temperature and radar echo. The precipitation types include rain, snow, freezing rain, ice rain, and hail; after the precipitation type is determined, quantitatively calculate the precipitation particle content of the model based on the radar reflectivity observation data to obtain the three-dimensional distribution of precipitation particles.

2. The method according to claim 1, wherein: The conventional observation data includes the real-time observation data of surface automatic stations, aircraft reports, and sounding stations; The remote sensing data includes the real-time observation data of radar reflectivity, wind profiler radar, FY4A retrieved temperature profile, and ground-based microwave radiometer temperature profile.

3. The method according to claim 1 or 2, characterized in that: Assimilate the conventional meteorological element observations into the background field through the Bratseth successive correction method. For the nth iteration, the corresponding iterative expression is as follows: Wherein, n is the number of iterations; is the n-th iteration value at lattice point i; is the (n - 1)-th iteration value at lattice point i; is the observed value of the k-th observation point around lattice point i; nobs is the total number of grid points; is the (n - 1)th estimated value at the observation point k, which is obtained by lattice interpolation; α is the weight coefficient, which is determined by the observational error variance, the background error variance, and the distance from the grid point to the observation; α ik is the weight coefficient of the observation value of the k-th observation point around the grid point i.

4. The method according to claim 3, characterized in that: After each iteration ends, perform balance adjustment based on preset balance constraint conditions; And, according to the preset number of iteration cycles, when the number of iterations reaches the aforementioned number of iteration cycles, obtain the update information of the real-time observation data, and introduce new real-time observation data for re-iteration.

5. The method according to claim 1 or 2, characterized in that, The calculation of the three-dimensional grid cloud amount includes the steps of: according to the relative humidity information of the model background field, calculating the three-dimensional background field cloud amount using a preset empirical relationship, and determining the three-dimensional background field cloud amount distribution; then, using the cloud observation data in the ground real-time observation data to correct the cloud base, cloud amount, and cloud thickness; subsequently, further adjusting the cloud amount using the model-gridded radar reflectivity observation data; When adjusting the model temperature and humidity fields, based on the wet adiabatic or non-adiabatic initialization method.

6. The method according to claim 5, wherein The specific steps of calculating the three-dimensional grid cloud amount are as follows: obtain the relative humidity information of the model background field, and calculate the cloud amount using the following empirical relationship according to the relative humidity information of the model background field to obtain the three-dimensional background field cloud amount CF, Among them, CF is cloud cover, and its value ranges from 0 to 1.0; RH is the relative humidity of the model background field; RH0 is a relative humidity threshold that varies with height; b is a preset empirical constant; Then, using the cloud observation data of the ground truth observation data, the cloud base, cloud cover, and cloud thickness are corrected by the Barnes interpolation method; if there is no ground observation data related to clouds, no correction is made; Combined with the cloud cover of the three-dimensional background field and the temperature of the initial field, a cloud top bright temperature is calculated; The cloud cover value is further adjusted according to the radar reflectivity observation data; among them, when the radar reflectivity of a grid point is greater than the set threshold and the radar echo is above the cloud base, the cloud cover value of this grid point is set to 1.

7. The method according to claim 5, characterized in that, When constructing and calculating the three-dimensional cloud field, a cloud water and cloud ice calculation unit, a precipitation type calculation unit, and a precipitation particle quantitative calculation unit are configured; The cloud water and cloud ice calculation is configured as follows: The cloud is divided into multiple layers from the cloud base to the cloud top, and the temperature, air pressure, and saturated water vapor pressure of each layer are calculated layer by layer to obtain the saturated water vapor mixing ratio. The temperature of each layer inside the cloud required in the calculation is calculated according to the saturated adiabatic vertical lapse rate, the air pressure of each layer is calculated according to the pressure-height formula, and the saturated water vapor pressure of each layer is calculated through the Clausius-Clapeyron equation; then the difference in the saturated water vapor mixing ratio between adjacent layers is calculated as the basic increment or decrement of cloud water and cloud ice. At this time, the cloud droplet particle mixing ratio ALWC of each layer is expressed as follows: ALWC k = ALWC k-1 + q vs_k-1 - q vs_k Among them, k is the number of cloud layers; ALWC k is the mixing ratio of cloud droplet particles in the k-th layer; ALWC k-1 is the mixing ratio of cloud droplet particles in the (k - 1)-th layer; q vs is the saturated water vapor mixing ratio, q vs_k is the saturated water vapor mixing ratio in the k-th layer, q vs_k-1 is the saturated water vapor mixing ratio in the (k - 1)-th layer; after obtaining the mixing ratio of cloud droplet particles in each layer, the cloud ice and cloud water content are separated according to the environmental temperature T. Among them, when the temperature is above the first preset temperature threshold, it is determined to be all cloud water; when the temperature is below the second preset temperature threshold, it is determined to be all cloud ice; when the temperature is between the first preset temperature threshold and the second preset temperature threshold, it is determined to be a mixture of cloud ice and cloud water. Among them, the weight of cloud water is 0.05 * (T - the second preset temperature threshold), and the remaining part is cloud ice. T represents the environmental temperature value; the precipitation type calculation unit is configured to: starting from the echo top of each grid body, when the wet-bulb temperature Tw of the echo top is lower than 0°C, it is determined that the precipitate starts to be snow, otherwise it is determined to be rain; when the snow falls into an area where the wet-bulb temperature is higher than 1.3°C, it is determined that the precipitate melts into rain; when the rain falls into a place where the wet-bulb temperature is lower than 0°C, it is determined that the precipitate condenses into freezing rain; when the freezing rain falls between the preset air pressures P1 and P2, it is determined that the precipitate becomes sleet. The air pressures P1 and P2 satisfy the expression , where T represents the environmental temperature value, |dP| represents the absolute value of the differential dP, and mb represents millibar, which is a unit of pressure; when the radar echo intensity exceeds a certain set threshold, the precipitation type is determined to be hail; The precipitation particle quantitative calculation unit is configured to: obtain radar reflectivity observation data; according to the radar reflectivity observation data, based on the empirical relationship between radar reflectivity and different types of precipitation particles, quantitatively calculate the precipitation particle content; among them, the total radar reflectivity includes the contributions of different precipitation particles such as rain, snow, and hail to the reflectivity factor.

8. The method according to claim 7, wherein Two calculation modes are configured in the precipitation particle quantitative calculation, including the first calculation mode and the second calculation mode; The first calculation mode is the Kessler mode based on the parameterization method of fitting observations. At this time, the calculation formula for the reflectivity factor of rain is: Z = a(ρ·q r ) b where Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10 log 10 Z; a and b are preset empirical constants; ρ is the air density, with the unit of kg / m 3 ; q r is the rain water mixing ratio, with the unit of g / kg; The calculation formulas for the reflectivity factors of snow and hail are: Z = c(ρ·q s ) d Among them, Z is the reflectivity factor, with the unit of mm 6 / m 3 , the radar reflectivity unit is dBZ, and 1 dBZ = 10 log 10 Z; c and d are preset empirical constants; ρ is the air density, with the unit of kg / m 3 ; q s is the rainwater mixing ratio, with the unit of g / kg; The second calculation mode is the Ferrier mode based on the cloud physical process and the backscattering method of hydrometeors. At this time, the relationship between the reflectivity factor equation and the model variables is: Z e = Z er + Z es + Z eh Among them, Z e is the total reflectivity factor, with the unit of mm 6 / m 3 , including the contributions of the rain reflectivity factor Z er , the snow reflectivity factor Z es , and the hail reflectivity factor Z eh ; For rain, its reflectivity factor Z er The equation expression is: Among them, ρ r represents the rainwater density, and ρ represents the air density; N r represents the intercept that satisfies the Marshall-Palmer drop size distribution; q r represents the rainwater mixing ratio; For snow, it is divided into dry snow and wet snow. When the temperature is lower than 0°C, it is defined as dry snow, and vice versa; Reflectivity factor Z of dry snow es The equation expression is as follows: Among them, ρ s represents the density of snow, ρ represents the air density, and ρ i represents the density of ice, N s represents the intercept of snow, K i 2 and K r 2 represent the dielectric constants of ice and water respectively, and q s represents the mixing ratio of dry snow; Reflectivity factor Z of wet snow es The equation expression is as follows: Among them, ρ s represents the density of snow, ρ represents the air density, N s represents the intercept of snow, q s represents the mixing ratio of wet snow; For hail, the reflectivity factor Z eh The equation expression is as follows: Among them, ρ h represents the density of hail, ρ represents the air density, N h represents the intercept of hail, q h represents the mixing ratio of hail.

9. A three-dimensional live analysis system according to the method of claim 1, characterized in that Including: A forecast field downscaling module, which is used to receive the three-dimensional forecast field data of the target area, and the horizontal resolution of the three-dimensional forecast field is the initial horizontal resolution; Perform downscaling processing on the aforementioned three-dimensional forecast field to obtain background field data with the target horizontal resolution; The value of the target horizontal resolution is less than the value of the initial horizontal resolution, so that the resolution of the background field is higher than that of the forecast field; An observation data collection and processing module, which is used to collect and organize multi-source observation data, and the multi-source observation data includes conventional observation data and remote sensing data; An assimilation analysis module, configured to receive multi-source observation data transmitted by an observation data collection and processing module, and fuse the multi-source observation data into high-resolution background field data to perform correction processing and cloud analysis on the background field, so as to obtain a high-resolution three-dimensional real-time analysis field; wherein, a successive correction method is adopted to assimilate conventional meteorological element observations into the background field to correct the model background field, and a complex cloud analysis method is combined with radar reflectivity observations to perform water substance phase analysis and in-cloud temperature and humidity field adjustment; A display module, configured to output the three-dimensional real-time analysis field described above.

10. A computer-readable storage medium, on which computer program instructions are stored, characterized in that: The computer program instructions, when executed by a processor, implement the method according to any one of claims 1-8.

Images