A method for assimilating GNSS multi-dimensional water vapor with WRF weather model

Abstract

The application discloses a kind of assimilation methods based on GNSS multidimensional water vapor cooperates WRF weather model, comprising: one, GNSS water vapor retrieval;Two, the first assimilation of precipitable water of GNSS retrieval;Three, GNSS three-dimensional tomography retrieval;Four, the water vapor density value of three-dimensional tomography grid is interpolated into forecast three-dimensional grid, obtain the water vapor density value of the same three-dimensional interpolation grid with forecast three-dimensional grid;Five, the relative humidity profile is retrieved by water vapor density value of three-dimensional interpolation grid and temperature data of forecast three-dimensional grid;Six, the second assimilation of relative humidity profile.The application design is reasonable, first, the precipitable water of GNSS retrieval is used to carry out the first assimilation, and the relative humidity profile is obtained using three-dimensional water vapor density and the temperature data of first assimilation, and the relative humidity profile is carried out the second assimilation based on WRF weather model, effectively utilize three-dimensional water vapor density data assimilation, improve weather parameter forecast precision.

Description

Technical Field

[0001] This invention belongs to the field of meteorological data assimilation technology, specifically relating to an assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model. Background Technology

[0002] Water vapor is a greenhouse gas in the atmosphere, playing a major driving role in weather occurrence and development. Precipitable water vapor (PWV), as one form of atmospheric water vapor content, is a significant factor influencing related weather processes. Global Navigation Satellite System (GNSS) water vapor detection technology is unaffected by weather conditions and offers advantages such as high precision, low cost, and continuous operation, providing high-precision, high-resolution, multi-dimensional water vapor information. Therefore, the assimilation of PWV retrieved from GNSS based on WRF meteorological models has emerged to improve the accuracy of meteorological parameter forecasts.

[0003] However, the current assimilation method uses PWV (Physical Water Vapor Volume), which only provides information on water vapor content per unit area in the zenith direction of the GNSS station, failing to comprehensively reflect the water vapor distribution in the surrounding space. Furthermore, the three-dimensional water vapor density data obtained based on GNSS tomography has not been fully utilized in the assimilation of WRF meteorological model data, limiting its potential to improve the accuracy of meteorological parameter forecasts.

[0004] Therefore, a well-designed assimilation method for GNSS multidimensional water vapor combined with WRF meteorological models is needed. First, the precipitable water is inverted from GNSS for initial assimilation. Then, the relative humidity profile is obtained using three-dimensional water vapor density and the temperature data from the initial assimilation. Finally, the relative humidity profile is assimilated again based on the WRF meteorological model, effectively utilizing three-dimensional water vapor density data for assimilation. This aims to improve the accuracy of the initial water vapor field of the WRF meteorological model and enhance the accuracy of meteorological parameter forecasts. Summary of the Invention

[0005] The technical problem to be solved by this invention is to address the shortcomings of the prior art by providing an assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model. The method has simple steps and a reasonable design. First, the precipitable water is inverted from GNSS for initial assimilation. Then, the relative humidity profile is obtained using the three-dimensional water vapor density and the temperature data from the initial assimilation. The relative humidity profile is then assimilated again based on the WRF meteorological model, effectively utilizing the three-dimensional water vapor density data for assimilation. This improves the accuracy of the initial water vapor field of the WRF meteorological model and enhances the accuracy of meteorological parameter forecasts.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: an assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model, characterized in that the method includes the following steps:

[0007] Step 1: GNSS water vapor inversion:

[0008] Step 101: Using a computer, the precipitable water volume PWV of the area under study is obtained by GNSS inversion according to the formula PWV = Π·ZWD; where ZWD represents the zenith wet delay and Π represents the atmospheric water vapor conversion coefficient.

[0009] Step 102: Repeat step 101 multiple times to obtain the precipitable water inversion from multiple GNSS stations; wherein, the precipitable water inversion from the j-th GNSS station is recorded as PWV(j) by computer, and the longitude, latitude and altitude corresponding to the j-th GNSS station are recorded as lat1(j), lon1(j) and h1(j) respectively.

[0010] Step 2: Initial assimilation of GNSS-retrieved precipitable water:

[0011] Step 201: Use a computer to convert the longitude lat1(j), latitude lon1(j), altitude h1(j), precipitable water volume PWV(j), and first error Δ1(j) corresponding to the j-th GNSS station into precipitable water volume data in Little_R format;

[0012] Step 202: Use a computer to input the Little_R format precipitable water data into the OBSPROC module of the WRF meteorological model observation processor to process it into an observation file for WRFDA.

[0013] Step 203: Use a computer to input the observation file for WRFDA into the WRFDA module of the WRF meteorological model for three-dimensional variational assimilation, and output a meteorological data file in NetCDF format.

[0014] Step 204: Extract temperature data from the forecast 3D grid from the meteorological data file in NetCDF format;

[0015] Step 3: GNSS 3D tomography inversion:

[0016] Step 301: Divide the study area into grids along the longitude, latitude, and altitude directions to form a three-dimensional tomographic grid; and use the GNSS water vapor tomography model to obtain the water vapor density inversion value of the three-dimensional tomographic grid.

[0017] Step 302: Using the trained LSTM neural network model, input the water vapor content and wet delay of the oblique path to obtain the predicted value of water vapor density of the three-dimensional tomographic grid.

[0018] Step 303: Average the water vapor density inversion value and the water vapor density prediction value of the three-dimensional tomographic grid to obtain the water vapor density value of the three-dimensional tomographic grid.

[0019] Step 4: Interpolate the water vapor density values ​​of the three-dimensional tomographic grid into the predicted three-dimensional grid to obtain the water vapor density values ​​of the three-dimensional interpolated grid that are the same as those of the predicted three-dimensional grid;

[0020] Step 5: Invert the water vapor density values ​​of the 3D interpolated grid with the temperature data of the predicted 3D grid to obtain the relative humidity profile:

[0021] Step 501: Obtain the water vapor density value ρ of the i′-th grid from the water vapor density values ​​of the three-dimensional interpolation grid. 0i′ Simultaneously, the temperature T of the i′-th grid is obtained from the temperature data of the predicted three-dimensional grid. i′ and saturated water vapor pressure e si′ Where i′ is a positive integer;

[0022] Step 502: Use a computer to calculate the formula. The vapor pressure of the i′-th grid is obtained; where R v M is the gas constant for water vapor; w Indicates the molar mass of water vapor;

[0023] Step 503, according to Obtain the relative humidity (RH) of the i′-th grid. i′ ;

[0024] Step 504: Repeat step 503 multiple times to obtain the relative humidity profile; wherein, the relative humidity profile includes the longitude, latitude, altitude of each grid and the relative humidity corresponding to each grid.

[0025] Step Six: Re-assimilation of the Relative Humidity Profile:

[0026] Step 601: Convert the longitude, latitude, altitude of each grid, the relative humidity of each grid, and the second error corresponding to each relative humidity into relative humidity data in Little_R format;

[0027] Step 602: Following the methods in steps 202 and 203, input the relative humidity data in Little_R format into the WRF meteorological model for three-dimensional variational assimilation again, and output a meteorological data file in NetCDF format.

[0028] The assimilation method based on the GNSS multidimensional water vapor co-located WRF meteorological model is characterized by the following process for obtaining the zenith wet delay (ZWD) and atmospheric water vapor conversion coefficient Π in step 101:

[0029] Step A: Use a computer and GAMIT / GLOBK software to solve the GNSS station data and obtain the total zenith delay (ZTD).

[0030] Step B: Using a computer to utilize the Saastamoinen model The zenith dry delay (ZHD) is obtained; where P represents the surface air pressure at the location of the GNSS station. H represents the latitude of the GNSS station, and H represents the geodetic height of the GNSS station.

[0031] Step C: Use a computer to obtain the zenith wet delay ZWD according to the formula ZWD = ZTD - ZHD;

[0032] Step D: Using a computer to... Obtain the weighted average temperature T m Where I represents the atmospheric layer number, i is the layer number, and both i and I are positive integers, and 1 ≤ i ≤ I; T i Let e ​​represent the absolute temperature at the top of the i-th layer of the atmosphere. i Let Δh represent the water vapor pressure at the top of the i-th layer of the atmosphere. i This represents the thickness of the i-th atmosphere;

[0033] Step E: Use a computer to calculate the formula. The atmospheric water vapor conversion coefficient Π is obtained; where ρ w R represents the density of liquid water. v T represents the water vapor gas constant. m K represents the weighted average temperature, k2' represents the first constant of atmospheric refractive index, and k3 represents the second constant of atmospheric refractive index.

[0034] The above-mentioned assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model is characterized by the following: the LSTM neural network model trained in step 302 is specifically implemented as follows:

[0035] Step 3021: Using the GNSS water vapor tomography model, obtain the water vapor density inversion values ​​of the three-dimensional tomography grid at each time step within the tomography model;

[0036] Step 3022: Repeat step 3021 multiple times to obtain the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, and use the water vapor content of the oblique path and the oblique path wet delay at multiple time points, as well as the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, as the training set.

[0037] Step 3023: Construct the LSTM neural network model;

[0038] Step 3024: Take the water vapor content and wet delay of the oblique path in the training set as inputs, and the water vapor density value of the three-dimensional tomographic grid as the output layer, and input them into the LSTM neural network model for training to obtain the trained LSTM neural network model.

[0039] The assimilation method based on the GNSS multidimensional water vapor co-located WRF meteorological model described above is characterized by: Step four, the specific process of which is as follows:

[0040] Step 401: Denote any intersection point of the predicted 3D grid as the target grid (lat). t ,lon t ,z t ), (lat t ,lon t ,z t The longitude, latitude, and altitude of the target grid are respectively;

[0041] Step 402: Obtain the latitude and longitude of the four adjacent grid centers on the lower horizontal plane projected downward from the target grid in the 3D tomographic grid, and denote the latitude and longitude of the first grid center as (lat1, lon1), the second grid center as (lat1, lon2), the third grid center as (lat2, lon1), and the fourth grid center as (lat2, lon2); where the height of the lower horizontal plane is z1;

[0042] The water vapor densities of the four adjacent grid centers are denoted as the water vapor density value ρ of the first grid center. 11 The water vapor density value ρ at the center of the second grid 12 The water vapor density value ρ at the center of the third grid 21 The water vapor density value ρ at the center of the fourth grid 22 ;

[0043] Step 403, according to ρ(lat) t ,lon t ,z1)=(1-α)(1-β)ρ 11 +α(1-β)ρ 21 +β(1-α)ρ 12 +αβρ 22 The water vapor density value ρ(lat) is obtained by projecting the target mesh downwards onto the lower horizontal plane at a height z1. t ,lon t ,z1); where α represents the first interpolation weight, and β represents the second interpolation weight, and

[0044]

[0045] Step 404: Following the methods in steps 402 and 403, obtain the water vapor density value ρ(lat) projected upwards onto the target mesh at a height z2 on the upper horizontal plane. t ,lon t ,z2);

[0046] Step 405, according to Obtain the water vapor density value ρ(lat) of the target grid. t ,lon t ,z t );

[0047] Step 405: Repeat steps 401 to 404 multiple times to interpolate the water vapor density value of the three-dimensional tomographic grid into the predicted three-dimensional grid, and obtain the water vapor density value of the three-dimensional interpolated grid that is the same as the predicted three-dimensional grid.

[0048] The assimilation method based on the GNSS multidimensional water vapor co-located WRF meteorological model described above is characterized by the following process for obtaining the first error Δ1(j) in step 201:

[0049] The precipitable water obtained by radiosonde is taken as the true value. The error between the precipitable water retrieved from the j-th GNSS station and the true value is obtained by computer and denoted as Δ1(j).

[0050] The specific process for obtaining the second error in step 601 is as follows:

[0051] The relative humidity obtained by the radiosonde is taken as the true value. The error between each relative humidity and the true value is obtained by computer and recorded as the second error.

[0052] Compared with the prior art, the present invention has the following advantages:

[0053] 1. This invention performs initial assimilation based on precipitable water retrieved from GNSS, and further assimilation based on relative humidity profiles. It uses GNSS multidimensional water vapor information to improve the relative humidity information of numerical models and enhance the forecast accuracy of the models.

[0054] 2. When interpolating the water vapor density values ​​of the three-dimensional tomographic grid into the prediction three-dimensional grid, this invention uses bilinear interpolation and height layer interpolation techniques to generate high-precision data that can be assimilated. Combined with temperature data, it is converted into an assimilated relative humidity profile, thus achieving effective assimilation of water vapor information.

[0055] 3. This invention combines the water vapor density inversion values ​​of the three-dimensional tomographic grid obtained from the GNSS water vapor tomography model and the water vapor density prediction values ​​of the three-dimensional tomographic grid obtained from the LSTM neural network model to obtain high-resolution three-dimensional water vapor information. This information is then converted to generate a directly assimilated relative humidity profile. Further assimilation using this relative humidity profile improves the initial field humidity information of the numerical weather prediction model. Through multi-dimensional water vapor data assimilation, the accuracy of the initial field humidity information of the numerical weather prediction model is improved, laying the foundation for accurate numerical forecasting of subsequent meteorological parameters.

[0056] 4. This invention overcomes the shortcomings of traditional two-dimensional assimilation in vertical spatial resolution, enabling the WRF meteorological model to capture the spatiotemporal distribution characteristics of water vapor in the atmosphere more comprehensively. In particular, the detailed characterization in the vertical direction is enhanced, thereby improving the WRF meteorological model's ability to simulate and forecast complex weather processes.

[0057] 5. This invention is compatible with existing WRF meteorological models and GNSS station observations, and has high universality and practicality, and can be widely applied to different regions and various types of meteorological forecasting tasks.

[0058] In summary, the method of this invention is simple in steps and reasonable in design. First, the precipitable water is assimilated using GNSS inversion. Then, the relative humidity profile is obtained using three-dimensional water vapor density and the temperature data from the first assimilation. Finally, the relative humidity profile is assimilated again based on the WRF meteorological model. This effectively utilizes three-dimensional water vapor density data for assimilation, thereby improving the accuracy of the initial water vapor field of the WRF meteorological model and increasing the accuracy of meteorological parameter forecasts.

[0059] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0060] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0061] like Figure 1 As shown, the assimilation method based on the GNSS multidimensional water vapor co-located WRF meteorological model of the present invention includes the following steps:

[0062] Step 1: GNSS water vapor inversion:

[0063] Step 101: Using a computer, the precipitable water volume PWV of the area under study is obtained by GNSS inversion according to the formula PWV = Π·ZWD; where ZWD represents the zenith wet delay and Π represents the atmospheric water vapor conversion coefficient.

[0064] Step 102: Repeat step 101 multiple times to obtain the precipitable water inversion from multiple GNSS stations; wherein, the precipitable water inversion from the j-th GNSS station is recorded as PWV(j) by computer, and the longitude, latitude and altitude corresponding to the j-th GNSS station are recorded as lat1(j), lon1(j) and h1(j) respectively.

[0065] Step 2: Initial assimilation of GNSS-retrieved precipitable water:

[0066] Step 201: Use a computer to convert the longitude lat1(j), latitude lon1(j), altitude h1(j), precipitable water volume PWV(j), and first error Δ1(j) corresponding to the j-th GNSS station into precipitable water volume data in Little_R format;

[0067] Step 202: Use a computer to input the Little_R format precipitable water data into the OBSPROC module of the WRF meteorological model observation processor to process it into an observation file for WRFDA.

[0068] Step 203: Use a computer to input the observation file for WRFDA into the WRFDA module of the WRF meteorological model for three-dimensional variational assimilation, and output a meteorological data file in NetCDF format.

[0069] Step 204: Extract temperature data from the forecast 3D grid from the meteorological data file in NetCDF format;

[0070] Step 3: GNSS 3D tomography inversion:

[0071] Step 301: Divide the study area into grids along the longitude, latitude, and altitude directions to form a three-dimensional tomographic grid; and use the GNSS water vapor tomography model to obtain the water vapor density inversion value of the three-dimensional tomographic grid.

[0072] Step 302: Using the trained LSTM neural network model, input the water vapor content and wet delay of the oblique path to obtain the predicted value of water vapor density of the three-dimensional tomographic grid.

[0073] Step 303: Average the water vapor density inversion value and the water vapor density prediction value of the three-dimensional tomographic grid to obtain the water vapor density value of the three-dimensional tomographic grid.

[0074] Step 4: Interpolate the water vapor density values ​​of the three-dimensional tomographic grid into the predicted three-dimensional grid to obtain the water vapor density values ​​of the three-dimensional interpolated grid that are the same as those of the predicted three-dimensional grid;

[0075] Step 5: Invert the water vapor density values ​​of the 3D interpolated grid with the temperature data of the predicted 3D grid to obtain the relative humidity profile:

[0076] Step 501: Obtain the water vapor density value ρ of the i′-th grid from the water vapor density values ​​of the three-dimensional interpolation grid. 0i′ Simultaneously, the temperature T of the i′-th grid is obtained from the temperature data of the predicted three-dimensional grid. i′ and saturated water vapor pressure e si′ Where i′ is a positive integer;

[0077] Step 502: Use a computer to calculate the formula. The vapor pressure of the i′-th grid is obtained; where R v M is the gas constant for water vapor; w Indicates the molar mass of water vapor;

[0078] Step 503, according to Obtain the relative humidity (RH) of the i′-th grid. i′ ;

[0079] Step 504: Repeat step 503 multiple times to obtain the relative humidity profile; wherein, the relative humidity profile includes the longitude, latitude, altitude of each grid and the relative humidity corresponding to each grid.

[0080] Step Six: Re-assimilation of the Relative Humidity Profile:

[0081] Step 601: Convert the longitude, latitude, altitude of each grid, the relative humidity of each grid, and the second error corresponding to each relative humidity into relative humidity data in Little_R format;

[0082] Step 602: Following the methods in steps 202 and 203, input the relative humidity data in Little_R format into the WRF meteorological model for three-dimensional variational assimilation again, and output a meteorological data file in NetCDF format.

[0083] In this embodiment, the specific process for obtaining the zenith wet delay (ZWD) and atmospheric water vapor conversion coefficient (Π) in step 101 is as follows:

[0084] Step A: Use a computer and GAMIT / GLOBK software to solve the GNSS station data and obtain the total zenith delay (ZTD).

[0085] Step B: Using a computer to utilize the Saastamoinen model The zenith dry delay (ZHD) is obtained; where P represents the surface air pressure at the location of the GNSS station. H represents the latitude of the GNSS station, and H represents the geodetic height of the GNSS station.

[0086] Step C: Use a computer to obtain the zenith wet delay ZWD according to the formula ZWD = ZTD - ZHD;

[0087] Step D: Using a computer to... Obtain the weighted average temperature T m Where I represents the atmospheric layer number, i is the layer number, and both i and I are positive integers, and 1 ≤ i ≤ I; T i Let e ​​represent the absolute temperature at the top of the i-th layer of the atmosphere. i Let Δh represent the water vapor pressure at the top of the i-th layer of the atmosphere. i This represents the thickness of the i-th atmosphere;

[0088] Step E: Use a computer to calculate the formula. The atmospheric water vapor conversion coefficient Π is obtained; where ρ w R represents the density of liquid water. v T represents the water vapor gas constant. m K represents the weighted average temperature, k2' represents the first constant of atmospheric refractive index, and k3 represents the second constant of atmospheric refractive index.

[0089] In this embodiment, the LSTM neural network model trained in step 302 is processed as follows:

[0090] Step 3021: Using the GNSS water vapor tomography model, obtain the water vapor density inversion values ​​of the three-dimensional tomography grid at each time step within the tomography model;

[0091] Step 3022: Repeat step 3021 multiple times to obtain the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, and use the water vapor content of the oblique path and the oblique path wet delay at multiple time points, as well as the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, as the training set.

[0092] Step 3023: Construct the LSTM neural network model;

[0093] Step 3024: Take the water vapor content and wet delay of the oblique path in the training set as inputs, and the water vapor density value of the three-dimensional tomographic grid as the output layer, and input them into the LSTM neural network model for training to obtain the trained LSTM neural network model.

[0094] In this embodiment, step four is as follows:

[0095] Step 401: Denote any intersection point of the predicted 3D grid as the target grid (lat). t ,lon t ,z t ), (lat t ,lon t ,z t The longitude, latitude, and altitude of the target grid are respectively;

[0096] Step 402: Obtain the latitude and longitude of the four adjacent grid centers on the lower horizontal plane projected downward from the target grid in the 3D tomographic grid, and denote the latitude and longitude of the first grid center as (lat1, lon1), the second grid center as (lat1, lon2), the third grid center as (lat2, lon1), and the fourth grid center as (lat2, lon2); where the height of the lower horizontal plane is z1;

[0097] The water vapor densities of the four adjacent grid centers are denoted as the water vapor density value ρ of the first grid center. 11 The water vapor density value ρ at the center of the second grid 12 The water vapor density value ρ at the center of the third grid 21 The water vapor density value ρ at the center of the fourth grid 22 ;

[0098] Step 403, according to ρ(lat) t ,lon t ,z1)=(1-α)(1-β)ρ 11 +α(1-β)ρ 21 +β(1-α)ρ 12 +αβρ 22 The water vapor density value ρ(lat) is obtained by projecting the target mesh downwards onto the lower horizontal plane at a height z1. t ,lon t ,z1); where α represents the first interpolation weight, and β represents the second interpolation weight, and

[0099]

[0100] Step 404: Following the methods in steps 402 and 403, obtain the water vapor density value ρ(lat) projected upwards onto the target mesh at a height z2 on the upper horizontal plane. t ,lon t ,z2);

[0101] Step 405, according to Obtain the water vapor density value ρ(lat) of the target grid. t ,lon t ,z t );

[0102] Step 405: Repeat steps 401 to 404 multiple times to interpolate the water vapor density value of the three-dimensional tomographic grid into the predicted three-dimensional grid, and obtain the water vapor density value of the three-dimensional interpolated grid that is the same as the predicted three-dimensional grid.

[0103] In this embodiment, the specific process for obtaining the first error Δ1(j) in step 201 is as follows:

[0104] The precipitable water obtained by radiosonde is taken as the true value. The error between the precipitable water retrieved from the j-th GNSS station and the true value is obtained by computer and denoted as Δ1(j).

[0105] The specific process for obtaining the second error in step 601 is as follows:

[0106] The relative humidity obtained by the radiosonde is taken as the true value. The error between each relative humidity and the true value is obtained by computer and recorded as the second error.

[0107] In this embodiment, specifically, the unit of the surface air pressure P corresponding to the GNSS station location in step B is hPa, and the latitude of the GNSS station is... The unit is rad, and the unit of GNSS station geodetic height H is km.

[0108] In this embodiment, when implementing the specific implementation, the value of I in step D is 37, and the specific stratification refers to the 37 pressure stratifications of the ERA5 upper-air ground data.

[0109] In this embodiment, specifically during implementation, the density ρ of liquid water in step E is... w The value is 1.0 × 10 3 kg / m 3 R v The value is 461.495 J / (kg·K), T m The unit is K, and the value of k2′ is 17±10 K / hPa, and the value of k3 is (3.776±0.04)×10 5 K 2 / hPa.

[0110] In this embodiment, the LSTM neural network model is configured with sigmoid and tanh activation functions, 2 neurons in the input layer, 500 neurons in the hidden layer, and the number of neurons in the output layer is the same as the number of three-dimensional tomographic grids. The learning rate is between 0.0001 and 0.001, and the number of training iterations is between 100 and 200.

[0111] In summary, the method of this invention is simple in steps and reasonable in design. First, the precipitable water is assimilated using GNSS inversion. Then, the relative humidity profile is obtained using three-dimensional water vapor density and the temperature data from the first assimilation. Finally, the relative humidity profile is assimilated again based on the WRF meteorological model. This effectively utilizes three-dimensional water vapor density data for assimilation, thereby improving the accuracy of the initial water vapor field of the WRF meteorological model and increasing the accuracy of meteorological parameter forecasts.

[0112] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model, characterized in that, The method includes the following steps: Step 1: GNSS water vapor inversion: Step 101: Using a computer, the precipitable water volume PWV of the area under study is obtained by GNSS inversion according to the formula PWV = Π·ZWD; where ZWD represents the zenith wet delay and Π represents the atmospheric water vapor conversion coefficient. Step 102: Repeat step 101 multiple times to obtain the precipitable water inversion from multiple GNSS stations; wherein, the precipitable water inversion from the j-th GNSS station is recorded as PWV(j) by computer, and the longitude, latitude and altitude corresponding to the j-th GNSS station are recorded as lat1(j), lon1(j) and h1(j) respectively. Step 2: Initial assimilation of GNSS-retrieved precipitable water: Step 201: Use a computer to convert the longitude lat1(j), latitude lon1(j), altitude h1(j), precipitable water volume PWV(j), and first error Δ1(j) corresponding to the j-th GNSS station into precipitable water volume data in Little_R format; Step 202: Use a computer to input the Little_R format precipitable water data into the OBSPROC module of the WRF meteorological model and process it into an observation file for WRFDA. Step 203: Use a computer to input the observation file for WRFDA into the WRFDA module of the WRF meteorological model for three-dimensional variational assimilation, and output a meteorological data file in NetCDF format. Step 204: Extract temperature data from the forecast 3D grid from the meteorological data file in NetCDF format; Step 3: GNSS 3D tomography inversion: Step 301: Divide the study area into grids along the longitude, latitude, and altitude directions to form a three-dimensional tomographic grid; and use the GNSS water vapor tomography model to obtain the water vapor density inversion value of the three-dimensional tomographic grid. Step 302: Using the trained LSTM neural network model, input the water vapor content and wet delay of the oblique path to obtain the predicted value of water vapor density of the three-dimensional tomographic grid. Step 303: Average the water vapor density inversion value and the water vapor density prediction value of the three-dimensional tomographic grid to obtain the water vapor density value of the three-dimensional tomographic grid. Step 4: Interpolate the water vapor density values ​​of the three-dimensional tomographic grid into the predicted three-dimensional grid to obtain the water vapor density values ​​of the three-dimensional interpolated grid that are the same as those of the predicted three-dimensional grid; Step 5: Invert the water vapor density values ​​of the 3D interpolated grid with the temperature data of the predicted 3D grid to obtain the relative humidity profile: Step 501: Obtain the water vapor density value ρ of the i′-th grid from the water vapor density values ​​of the three-dimensional interpolation grid. 0i′ Simultaneously, the temperature T of the i′-th grid is obtained from the temperature data of the predicted three-dimensional grid. i′ and saturated water vapor pressure e si′ Where i′ is a positive integer; Step 502: Use a computer to calculate the formula. The vapor pressure of the i′-th grid is obtained; where R v M is the gas constant for water vapor; w Indicates the molar mass of water vapor; Step 503, according to Obtain the relative humidity (RH) of the i′-th grid. i′ ; Step 504: Repeat step 503 multiple times to obtain the relative humidity profile; wherein, the relative humidity profile includes the longitude, latitude, altitude of each grid and the relative humidity corresponding to each grid. Step Six: Re-assimilation of the Relative Humidity Profile: Step 601: Convert the longitude, latitude, altitude of each grid, the relative humidity of each grid, and the second error corresponding to each relative humidity into relative humidity data in Little_R format; Step 602: Following the methods in steps 202 and 203, input the relative humidity data in Little_R format into the WRF meteorological model for three-dimensional variational assimilation again, and output a meteorological data file in NetCDF format.

2. The assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model according to claim 1, characterized in that: The specific process for obtaining the zenith wet delay (ZWD) and atmospheric water vapor conversion coefficient Π in step 101 is as follows: Step A: Use a computer and GAMIT / GLOBK software to calculate the GNSS station data and obtain the total zenith delay (ZTD). Step B: Using a computer to utilize the Saastamoinen model The zenith dry delay (ZHD) is obtained; where P represents the surface air pressure at the location of the GNSS station. H represents the latitude of the GNSS station, and H represents the geodetic height of the GNSS station. Step C: Use a computer to obtain the zenith wet delay ZWD according to the formula ZWD = ZTD - ZHD; Step D: Using a computer to... Obtain the weighted average temperature T m Where I represents the atmospheric layer number, i is the layer number, and both i and I are positive integers, and 1 ≤ i ≤ I; T i Let e ​​represent the absolute temperature at the top of the i-th layer of the atmosphere. i Let Δh represent the water vapor pressure at the top of the i-th layer of the atmosphere. i This represents the thickness of the i-th atmosphere; Step E: Use a computer to calculate the formula. The atmospheric water vapor conversion coefficient Π is obtained; where ρ w R represents the density of liquid water. v T represents the water vapor gas constant. m K represents the weighted average temperature, k2' represents the first constant of atmospheric refractive index, and k3 represents the second constant of atmospheric refractive index.

3. The assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model according to claim 1, characterized in that: The LSTM neural network model trained in step 302 is as follows: Step 3021: Using the GNSS water vapor tomography model, obtain the water vapor density inversion values ​​of the three-dimensional tomography grid at each time step within the tomography model; Step 3022: Repeat step 3021 multiple times to obtain the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, and use the water vapor content of the oblique path and the oblique path wet delay at multiple time points, as well as the water vapor density inversion values ​​of the three-dimensional tomographic grid at multiple time points, as the training set. Step 3023: Construct the LSTM neural network model; Step 3024: Take the water vapor content and wet delay of the oblique path in the training set as inputs, and the water vapor density value of the three-dimensional tomographic grid as the output layer, and input them into the LSTM neural network model for training to obtain the trained LSTM neural network model.

4. The assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model according to claim 1, characterized in that: Step four, the specific process is as follows: Step 401: Denote any intersection point of the predicted 3D grid as the target grid (lat). t ,lon t ,z t ), (lat t ,lon t ,z t The longitude, latitude, and altitude of the target grid are respectively; Step 402: Obtain the latitude and longitude of the four adjacent grid centers on the lower horizontal plane projected downward from the target grid in the 3D tomographic grid, and denote the latitude and longitude of the first grid center as (lat1, lon1), the second grid center as (lat1, lon2), the third grid center as (lat2, lon1), and the fourth grid center as (lat2, lon2); where the height of the lower horizontal plane is z1; The water vapor density at the four adjacent grid centers is denoted as ρ, which is the water vapor density value at the center of the first grid. 11 The water vapor density value ρ at the center of the second grid 12 The water vapor density value ρ at the center of the third grid 21 The water vapor density value ρ at the center of the fourth grid 22 ; Step 403, according to ρ(lat) t ,lon t ,z1)=(1-α)(1-β)ρ 11 +α(1-β)ρ 21 +β(1-α)ρ 12 +αβρ 22 The water vapor density value ρ(lat) is obtained by projecting the target mesh downwards onto the lower horizontal plane at a height z1. t ,lon t ,z1); where α represents the first interpolation weight, and β represents the second interpolation weight, and Step 404: Following the methods in steps 402 and 403, obtain the water vapor density value ρ(lat) projected upwards onto the target mesh at a height z2 on the upper horizontal plane. t ,lon t ,z2); Step 405, according to Obtain the water vapor density value ρ(lat) of the target grid. t ,lon t ,z t ); Step 405: Repeat steps 401 to 404 multiple times to interpolate the water vapor density value of the three-dimensional tomographic grid into the predicted three-dimensional grid, and obtain the water vapor density value of the three-dimensional interpolated grid that is the same as the predicted three-dimensional grid.

5. The assimilation method based on a GNSS multidimensional water vapor co-located WRF meteorological model according to claim 1, characterized in that: The specific process for obtaining the first error Δ1(j) in step 201 is as follows: The precipitable water obtained by radiosonde is taken as the true value. The error between the precipitable water retrieved from the j-th GNSS station and the true value is obtained by computer and denoted as Δ1(j). The specific process for obtaining the second error in step 601 is as follows: The relative humidity obtained by the radiosonde is taken as the true value. The error between each relative humidity and the true value is obtained by computer and recorded as the second error.

Images