CN117574748A - Plateau vortex initialization method with balanced power and hydrothermal structure - Google Patents
Plateau vortex initialization method with balanced power and hydrothermal structure Download PDFInfo
- Publication number
- CN117574748A CN117574748A CN202311293641.6A CN202311293641A CN117574748A CN 117574748 A CN117574748 A CN 117574748A CN 202311293641 A CN202311293641 A CN 202311293641A CN 117574748 A CN117574748 A CN 117574748A
- Authority
- CN
- China
- Prior art keywords
- vortex
- field
- plateau
- hydrothermal
- dimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011423 initialization method Methods 0.000 title abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000010801 machine learning Methods 0.000 claims abstract description 13
- 238000001556 precipitation Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 23
- 238000013527 convolutional neural network Methods 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 9
- 238000003908 quality control method Methods 0.000 claims description 7
- 241000208818 Helianthus Species 0.000 claims description 3
- 235000003222 Helianthus annuus Nutrition 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- JTJMJGYZQZDUJJ-UHFFFAOYSA-N phencyclidine Chemical class C1CCCCN1C1(C=2C=CC=CC=2)CCCCC1 JTJMJGYZQZDUJJ-UHFFFAOYSA-N 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 210000000746 body region Anatomy 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Fluid Mechanics (AREA)
- Algebra (AREA)
- Wind Motors (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a plateau vortex initialization method with balanced power and hydrothermal structure, which comprises the following steps: the method comprises the following steps of objectively identifying initial vortex: firstly extracting characteristic points of a plateau vortex; then clustering the characteristic points of the plateau vortex, and determining the center and the range of the plateau vortex; extracting a plateau vortex power field; removing the initial dynamic vortex to obtain a background field of the unpowered vortex; initializing a new vortex: reconstructing a vortex three-dimensional power structure, obtaining a vortex three-dimensional hydrothermal structure by using machine learning, and assimilating the vortex three-dimensional power structure obtained by reconstruction and the vortex three-dimensional hydrothermal structure obtained by machine learning as observation data to obtain a new vortex with the coordination of an initial moment dynamic field/structure and a hydrothermal field/structure. The invention can improve the prediction level of the altitude vortex and the precipitation thereof.
Description
Technical Field
The invention relates to the technical field of weather forecast, in particular to a plateau vortex initialization method with balanced power and hydrothermal structure.
Background
The low vortex of Qinghai-Tibet plateau (hereinafter referred to as Gao Yuanguo) is a mesoscale weather system which is active in the Qinghai-Tibet plateau main body region in summer and has a life history of about 2 to 3 days, and has a horizontal scale of about 500km and a vertical thickness of 2 to 3km. The plateau vortex not only affects the main precipitation system of the Qinghai-Tibet plateau in summer and the surrounding areas, but also can move out of the plateau in east under favorable weather conditions, thereby causing strong precipitation in the eastern area of China. At present, great deviation exists in forecasting of the plateau vortex and the rainfall intensity, the structure, the path and the like of the plateau vortex, so that the numerical forecasting level of the plateau vortex is improved, and the method is not only a research hotspot of Tibet plateau meteorology, but also one of difficulties of weather forecasting business.
Numerical forecasting is an initial value problem, and the error of an initial field is an important source of forecasting errors of the altitude vortex and the rainfall intensity, the path, the structure and the like of the altitude vortex. At present, the altitude vortex initialization technology mainly comprises 2 types: firstly, directly using the output field of a global mode with higher quality to downscale to obtain an initial field of a mesoscale mode; secondly, assimilating observation data of Qinghai-Tibet plateau areas on the basis of scale reduction so as to improve the initial field precision. However, both of these initialization techniques have a large limitation, and the prediction of plateau vortexes and precipitation thereof is limited due to the following main reasons: (1) Even the best quality global model product has larger deviation of the aerodynamic structure in the Qinghai-Tibet plateau area, especially in the main generation area of plateau vortex-the plateau midwestern area; (2) Studies on tropical cyclone active in ocean (ocean surface) show that a large error exists in a vortex power structure in a global mode, and a commonly used vortex reconstruction technology is to introduce axisymmetric or non-axisymmetric Rankine vortex (ideal vortex) to correct the vortex in the global mode; the altitude vortex is active in a mid-latitude land area, the inclined pressure process and structure are far more complex than a tropical cyclone on the ocean, the global mode and the existing vortex reconstruction technology have lower depicting capability on the altitude vortex which is a mesoscale vortex system power structure, so that the altitude vortex structure is inconsistent with an observed altitude vortex spiral rain belt structure, and a new method is needed to correct the altitude vortex power structure in the global mode; (3) The sounding and ground data of the Qinghai-Tibet plateau station are very limited, especially in the western part of the plateau, and the quality of the satellite, radar and other unconventional atmospheric profile data is lower under the condition of the Qinghai-Tibet plateau and the non-clear sky, so that the initialization level of the plateau vortex cannot be effectively improved due to the lack of effective input data of the plateau vortex when the existing data assimilation technology is simply used. Therefore, how to construct a more real vortex structure of the plateau vortex with coordinated power and water heat in the initial field of the numerical mode is a key for improving the plateau vortex and the rainfall forecast level thereof.
Representative methods of constructing a scroll power structure in an initial field at home and abroad are: iwasaki et al (1987) and Mathur (1991) propose a simple approach to vortex initialization, namely constructing an axisymmetric rank vortex as a function of maximum wind speed at the background field vortex location, which is widely used in the initialization of tropical cyclones (also known as "typhoons") on the ocean surface, which can significantly enhance the initial strength of the tropical cyclone. But it is disadvantageous in that: the dynamic structure of the new vortex is coarser, and has a certain difference with the real vortex; an imbalance of the wind field and the mass field may result. Kurihara et al (1992) introduced an asymmetric portion in an axisymmetric Rankin vortex based on the vortex initialization concept described above, the asymmetric portion being generated by a simplified positive pressure vorticity equation. The method has the advantage that the wind field and the mass field of the tropical cyclone on the ocean surface are balanced through a diagnosis equation, so that the new vortex is more real. Wu Yunfan et al (2021) improve the wind field structure of tropical cyclone vortices on the ocean surface based on empirical functions obtained from observations while taking into account the vortex kernel and peripheral wind ring information.
Currently, the existing vortex initialization method basically focuses on tropical cyclones (typhoons) active on the ocean (ocean surface), focuses on the dynamic structure of initial vortex, and neglects the adjustment of the vortex water and thermal structure. For mesoscale vortices (e.g. Gao Yuanguo) to be active on land, current initialisation studies still need to be further in depth. Moreover, the inclined pressure process is far more remarkable than the tropical cyclone (mainly represented as positive pressure process) on the ocean for the vortex on the land, so that the initialization of a vortex hydro-thermal structure is considered, and the initialization is coordinated with a vortex power structure, which is the key for improving the strength, the path and the rainfall forecast in the later vortex evolution process. In fact, the lack of adjustment or balancing of the hydrothermal structure in the initial vortex configuration active with tropical cyclones also leads to the problem of the new vortex not matching the mode background field (Low-Nam and Davis,2001;Kurihara et al, 1990).
In recent years, machine Learning (Machine Learning) technology has been widely used in weather analysis, weather forecast, climate forecast, and the like. For example, liu et al (2016) detect extreme weather events in climate products using Convolutional Neural Networks (CNNs); shi et al (2017) improved the level of short-run precipitation forecast using a method based on a combination of CNN and a circulation network; badrinath et al (2023) uses WRF precipitation prediction results as training data for Convolutional Neural Networks (CNNs) to significantly reduce the root mean square error of the daily accumulated precipitation on the west coast of the united states. As can be seen, machine learning methods can better improve the predictive power of areas of poor observational data, but these studies have focused on areas of flat terrain and rich atmospheric profile data. The Tibet plateau has complex terrain, less available atmospheric profile data, relatively reliable cloud top brightness and ground precipitation data observed by satellites, and further intensive research is needed to train and obtain a plateau vortex hydrothermal structure (field) by combining the satellite cloud top brightness and the precipitation data.
In summary, the main drawbacks and problems of the plateau vortex initialization method are:
(1) The conventional sounding and ground observation data of the Qinghai-Tibet plateau are insufficient, and the quality of the atmospheric profile data obtained by the irregular observation of satellites, radars and the like is lower, especially in the mid-west region of the plateau generated by plateau vortex. When the existing data assimilation technology is simply used, the initialization level of the plateau vortex is lower due to the lack of effective observation data input of the plateau vortex.
(2) When the global mode output field is directly used as an initial field of the plateau vortex, the global mode has larger error in the Qinghai-Tibet plateau, and the initialized plateau vortex power structure and hydrothermal structure have larger deviation.
(3) Since plateau vortex is active in mid-latitude land areas, there is a significant difference between the inclined pressure process and structure and tropical cyclone (typhoon) active on the ocean, the existing vortex initialization method for tropical cyclone is not suitable for initializing plateau vortex.
Disclosure of Invention
The invention aims to solve the technical problem of providing a plateau vortex initialization method for improving the balance between the power of a plateau vortex and the rainfall forecast level thereof and a hydrothermal structure.
In order to solve the problems, the plateau vortex initialization method with balanced power and hydrothermal structure comprises the following steps:
(1) Objective identification of initial vortex:
in the range covering Qinghai-Tibet plateau, the potential height field phi and horizontal wind field on 500hPa barometric layer of the global model forecast productSmoothing and extracting characteristic points of the plateau vortex; then clustering the characteristic points of the plateau vortex, and determining the center and the range of the plateau vortex;
(2) And (3) extracting a plateau vortex power field:
extracting Gao Yuanguo vortex force fields by using wind field and potential height field data on different air pressure layers on the basis of the determined altitude vortex center and altitude vortex range;
initial power vortex removal:
the background field of the unpowered vortex is obtained as follows:
and phi-phi'
Wherein:is a horizontal wind field, unit m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi is potential height field, unit m; />For rotating wind, units: m s -1 ；/>As a wind of dispersion, unit: m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi' is the disturbance potential height field, i.e. the difference between the vortex field and the background field, in units of: m;
initializing a new vortex:
reconstructing a vortex three-dimensional power structure, obtaining a vortex three-dimensional hydrothermal structure by using machine learning, and assimilating the vortex three-dimensional power structure obtained by reconstruction and the vortex three-dimensional hydrothermal structure obtained by machine learning as observation data to obtain a new vortex with the coordination of an initial moment dynamic field/structure and a hydrothermal field/structure.
The altitude vortex power field extraction in the step (2) is carried out according to the following method:
(1) reading in horizontal wind fieldCalculating relative vorticity zeta and divergence D;
(2) reading in potential height field phi, calculating relative vorticity zeta g ；
(3) Calculating rotational wind related to plateau vortexWind of degree of divergence->And a disturbing potential height field phi':
setting the zeta value of the relative vorticity outside the Gao Yuanguo range determined in the step (1) as 0, defining the Dirichlet boundary condition, namely the flow function as 0, and solving the following poisson equation to obtain the flow function psi on all air pressure layers;
ii calculating swirl-related rotational wind using flow function ψ
Wherein:is a unit vector in the vertical direction;
iii setting the value of the divergence D outside the Gao Yuanguo range determined in the step (1) to 0, defining the Dirichlet boundary condition, namely the potential function is 0, and solving the following Poisson equation to obtain the potential function χ on all air pressure layers:
iv calculating the wind with the divergence related to the vortex by using the potential function χ
V rotating the relative vorticity ζ of the ground outside the Gao Yuanguo range determined in step (1) g Set to 0, solve the following poisson equation to obtain the disturbance potential height field phi' associated with the vortex:
the reconstructed vortex three-dimensional power structure in the step IV is obtained by the following method: firstly, searching at least more than 3 sounding sites on the basis of the altitude vortex range determined in the step (1), and constructing a horizontal wind field by a curved surface fitting method; then calculating the relative vorticity, comparing with the relative vorticity of the background field, performing enhanced or weakened aggregate disturbance on the dynamic field of the plateau vortex, and introducing the aggregate-averaged dynamic field of the plateau vortex into the background field of the unpowered vortex, thereby constructing a new vortex dynamic fieldFinally, a new vortex power field is added>And performing quality control to obtain the product.
The vortex three-dimensional hydrothermal structure in the step III is obtained by the following method: selecting the temperature and dew point vertical profile of the sounding observation of the Qinghai-Tibet plateau station as output; cloud top bright temperature observed by a cloud satellite or a sunflower 8 satellite, and precipitation observed by GPM or CMORPM are used as input; and training to obtain the relationship between the cloud top bright temperature and the rainfall of the Qinghai-Tibet plateau and the atmospheric temperature and the dew point profile by using a convolutional neural network algorithm, and obtaining the initial time Gao Yuanguo vortex three-dimensional hydrothermal structure constructed based on the cloud top bright and the rainfall information.
The step of plateau vortex initialization is to take a vortex three-dimensional hydro-thermal structure and a reconstructed vortex three-dimensional power structure as observation data, assimilate the observation data by using a three-dimensional variation assimilation technology after quality control, and enter a numerical forecasting mode, so that a new vortex with a coordinated initial moment dynamic field/structure and a hydro-thermal field/structure is formed.
Compared with the prior art, the invention has the following advantages:
1. the method identifies, separates and removes the altitude vortex force field from the numerical mode initial field to form a background field of unpowered vortex; on the basis, a new vortex power field is introduced to obtain a plateau vortex power structure, and a convolutional neural network method is used for learning to obtain a plateau vortex hydrothermal structure; and finally, assimilating the altitude vortex power and the hydrothermal structure into an initial field by using a three-dimensional variation method to form a new vortex with a dynamic field coordinated with the hydrothermal field, so that the problem of insufficient atmospheric profile data in the current altitude vortex generation evolution research is solved.
2. The invention adopts a method of combining a reconstructed vortex three-dimensional power structure (field) and a machine learning vortex three-dimensional hydro-thermal structure (field) to coordinate an initial plateau vortex force field with a hydro-thermal field, and the plateau vortex in the mode initial field is more real. The method is unique in theory and practice, can fill the blank in the field of plateau vortex forecasting, and improves the forecasting level of plateau vortex and precipitation thereof.
3. The invention provides a new thought and approach for improving the level of the plateau vortex and the rainfall forecast thereof, and can be applied to rainfall forecast, land mesoscale vortex initialization, aviation weather, agricultural weather and disaster early warning.
Drawings
The following describes the embodiments of the present invention in further detail with reference to the drawings.
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a graph of flow function and wind field after removal of Gao Yuanguo vortex (b) and addition of new plateau vortex (c) using the present invention for ERA5 background field flow function (fill color) and wind field (wind vector) (a) in accordance with the present invention; and ERA5 background field potential height (color filling) (d), the potential height map after Gao Yuanguo vortex (e) and new plateau vortex (f) are removed by adopting the invention. Examples:
Detailed Description
As shown in fig. 1, a plateau vortex initialization method based on a reconstructed vortex power structure and a machine learning vortex hydrothermal structure comprises the following steps:
(1) Objective identification of initial vortex:
in the range covering Qinghai-Tibet plateau, the potential height field phi and horizontal wind field on 500hPa barometric layer of global model forecast product (such as 0.25 degree x 0.25 degree data of ERA 5)Smoothing and extracting characteristic points of the plateau vortex; and then clustering the characteristic points of the plateau vortex, and determining the center and the range of the plateau vortex. The plateau vortex range is taken as the vortex activity area.
Wherein: the condition of meeting the characteristic points of the plateau vortex means: the average value of the wind speeds of the south side latitude in the range of 5 degrees multiplied by 5 degrees around a certain grid point is larger than 0, the average value of the wind speeds of the north side latitude is smaller than 0, the average value of the warp direction wind speed of the east side is larger than 0, the average value of the warp direction wind speed of the west side is smaller than 0, and the wind direction meets the anticlockwise cyclone rotation; and the potential height value of the lattice point is smaller than the potential height average value in the range of 5 degrees multiplied by 5 degrees around the lattice point.
The clustering method adopts a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm, and the algorithm sets two parameters: the neighborhood radius and the minimum domain number are repeatedly optimized for the data of 0.25 degree multiplied by 0.25 degree of ERA5, and 1 and 10 are respectively taken.
(2) And (3) extracting a plateau vortex power field:
based on the determined altitude vortex center and altitude vortex range, gao Yuanguo vortex force field is extracted by using wind field and potential altitude field data on different air pressure layers (1000-100 hPa).
The specific process is as follows:
(1) reading in horizontal wind field(, unit m s) -1 ) The relative vorticity ζ (, units s) was calculated -1 ) Divergence D (, s) -1 )：
Wherein: u and v represent weft direction wind and warp direction wind respectively, unit m s -1 ；
(2) Reading in potential height field phi (unit m), calculating relative vorticity zeta g (Unit s) -1 )；
Wherein: f (f) 0 Representing the Coriolis force parameter, take 10 -4 ；
(3) Calculating rotational wind related to plateau vortexWind of degree of divergence->And a disturbing potential height field phi':
setting the value of the relative vorticity ζ outside the Gao Yuanguo range determined in the step (1) to 0, defining the dirichlet boundary condition, namely the flow function to 0, and solving the poisson equation to obtain the flow function ψ (, unit m) on all air pressure layers 2 s -1 )；
Ii calculating swirl-related rotational wind using flow function ψ
Wherein:is a unit vector in the vertical direction;
iii setting the value of the divergence D outside the Gao Yuanguo range determined in the step (1) to 0, defining the Dirichlet boundary condition, namely the potential function to be 0, and solving the following Poisson equation to obtain the potential function χ (, unit m) on all air pressure layers 2 s -1 )：
Iv calculating the wind with the divergence related to the vortex by using the potential function χ
V rotating the relative vorticity ζ of the ground outside the Gao Yuanguo range determined in step (1) g Set to 0, solve the poisson equation below to obtain the disturbance potential height field phi' (, unit m) associated with the vortex:
initial power vortex removal:
the background field of the unpowered vortex is obtained as follows:
and phi-phi'
Wherein:is a horizontal wind field, unit m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi is potential height field, unit m; />For rotating wind, units: m s -1 ；/>As a wind of dispersion, unit: m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi' is the disturbance potential height field, i.e. the difference between the vortex field and the background field, in units of: m.
Initializing a new vortex:
reconstructing a vortex three-dimensional power structure, obtaining a vortex three-dimensional hydrothermal structure by using machine learning, and assimilating the vortex three-dimensional power structure obtained by reconstruction and the vortex three-dimensional hydrothermal structure obtained by machine learning as observation data to obtain a new vortex with the coordination of an initial moment dynamic field/structure and a hydrothermal field/structure.
Wherein: the reconstructed vortex three-dimensional power structure is obtained by the following steps: firstly, searching at least more than 3 sounding sites on the basis of the altitude vortex range determined in the step (1), and constructing a horizontal wind field by a curved surface fitting method; the relative vorticity is then calculated and compared to the background field relative vorticity for the plateau vortex dynamic field (rotational windWind of degree of divergence->And the disturbance potential height field phi') is enhanced or weakened, and the concentrated average plateau vortex dynamic field is introduced into the background field of the unpowered vortex, so that a new vortex dynamic field is constructed>Finally, a new vortex power field is added>And 5 degrees of sparsification is carried out, the data is written into a Little_r data format as sounding data, and quality control is carried out, so that the sounding data is obtained.
The vortex three-dimensional hydrothermal structure is obtained by the following method: selecting the temperature and dew point vertical profile of the sounding observation of the Qinghai-Tibet plateau station as output (dependent variable, data sample over 5 years); cloud top bright temperature observed by a cloud satellite or a sunflower 8 satellite, and precipitation observed by GPM or CMORPM are taken as input (independent variable, more than 5 years of data samples). The temporal and spatial resolutions of the different data are the same and quality control is performed. And training to obtain the relationship between the bright temperature and the rainfall of the cloud top and the atmospheric temperature of the Qinghai-Tibet plateau and the dew point profile by using a convolutional neural network algorithm (Convolutional Neural Networks, CNN), and obtaining the initial time Gao Yuanguo vortex three-dimensional hydrothermal structure constructed based on the bright temperature and the rainfall information of the cloud top.
The initialization of the plateau vortex is to take the vortex three-dimensional hydro-thermal structure and the reconstructed vortex three-dimensional dynamic structure as observation data, assimilate the observation data by using a three-dimensional variation assimilation (3 DVAR) technology after quality control, and enter a numerical forecasting mode (such as a WRF mode), so that a new vortex with a coordinated initial moment dynamic field/structure and a hydro-thermal field/structure is formed.
Examples
Taking the origin-earth generation-elimination type altitude vortex occurring on day 8 and 14 of 2006 as an example, as shown in fig. 2. FIG. 2 shows (a-c) flow functions (filled) and wind farm (wind vector), (d-f) potential height (filled) and wind farm (wind vector) of 500hPa at 8.month, 14.day, 00. Wherein: (a, d) original background field, (b, e) removing background field after vortex, (c, f) adding new background field after vortex.
As can be seen from FIG. 2, the method of the present invention can correctly identify the position and the range of the ERA5 background field plateau vortex, remove the plateau vortex to obtain an unpowered vortex background field (FIG. 2b, e), and introduce a new vortex with the dynamic field balanced with the hydrothermal field (FIG. 2c, f).
Claims (5)
1. A method for initializing plateau vortex with balanced power and hydrothermal structure comprises the following steps:
(1) Objective identification of initial vortex:
in the range covering Qinghai-Tibet plateau, the potential height field phi and horizontal wind field on 500hPa barometric layer of the global model forecast productSmoothing and extracting characteristic points of the plateau vortex; then clustering the characteristic points of the plateau vortex, and determining the center and the range of the plateau vortex;
(2) And (3) extracting a plateau vortex power field:
extracting Gao Yuanguo vortex force fields by using wind field and potential height field data on different air pressure layers on the basis of the altitude vortex center and altitude vortex range determined in the step (1);
initial power vortex removal:
the background field of the unpowered vortex is obtained as follows:
and phi-phi'
Wherein:is a horizontal wind field, unit m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi is potential height field, unit m; />For rotating wind, units: m s -1 ；/>As a wind of dispersion, unit: m s -1 The method comprises the steps of carrying out a first treatment on the surface of the Phi' is the disturbance potential height field, i.e. the difference between the vortex field and the background field, in units of: m;
initializing a new vortex:
reconstructing a vortex three-dimensional power structure, obtaining a vortex three-dimensional hydrothermal structure by using machine learning, and assimilating the vortex three-dimensional power structure obtained by reconstruction and the vortex three-dimensional hydrothermal structure obtained by machine learning as observation data to obtain a new vortex with the coordination of an initial moment dynamic field/structure and a hydrothermal field/structure.
2. A method of initializing a plateau vortex with balanced power and hydrothermal structure as claimed in claim 1, wherein: the altitude vortex power field extraction in the step (2) is carried out according to the following method:
(1) reading in horizontal wind fieldCalculating relative vorticity zeta and divergence D;
(2) reading in potential height field phi, calculating relative vorticity zeta g ；
(3) Calculating rotational wind related to plateau vortexWind of degree of divergence->And a disturbing potential height field phi':
setting the zeta value of the relative vorticity outside the Gao Yuanguo range determined in the step (1) as 0, defining the Dirichlet boundary condition, namely the flow function as 0, and solving the following poisson equation to obtain the flow function psi on all air pressure layers;
ii calculating swirl-related rotational wind using flow function ψ
Wherein:is a unit vector in the vertical direction;
iii setting the value of the divergence D outside the Gao Yuanguo range determined in the step (1) to 0, defining the Dirichlet boundary condition, namely the potential function is 0, and solving the following Poisson equation to obtain the potential function χ on all air pressure layers:
iv calculating the wind with the divergence related to the vortex by using the potential function χ
V rotating the relative vorticity ζ of the ground outside the Gao Yuanguo range determined in step (1) g Set to 0, solve the following poisson equation to obtain the disturbance potential height field phi' associated with the vortex:
3. a method of initializing a plateau vortex with balanced power and hydrothermal structure as claimed in claim 1, wherein: the reconstructed vortex three-dimensional power structure in the step IV is obtained by the following method: firstly, searching at least more than 3 sounding sites on the basis of the altitude vortex range determined in the step (1), and constructing a horizontal wind field by a curved surface fitting method; then calculating the relative vorticity, comparing with the relative vorticity of the background field, performing enhanced or weakened aggregate disturbance on the dynamic field of the plateau vortex, and introducing the aggregate-averaged dynamic field of the plateau vortex into the background field of the unpowered vortex, thereby constructing a new vortex dynamic fieldFinally, a new vortex power field is added>And performing quality control to obtain the product.
4. A method of initializing a plateau vortex with balanced power and hydrothermal structure as claimed in claim 1, wherein: the vortex three-dimensional hydrothermal structure in the step III is obtained by the following method: selecting the temperature and dew point vertical profile of the sounding observation of the Qinghai-Tibet plateau station as output; cloud top bright temperature observed by a cloud satellite or a sunflower 8 satellite, and precipitation observed by GPM or CMORPM are used as input; and training to obtain the relationship between the cloud top bright temperature and the rainfall of the Qinghai-Tibet plateau and the atmospheric temperature and the dew point profile by using a convolutional neural network algorithm, and obtaining the initial time Gao Yuanguo vortex three-dimensional hydrothermal structure constructed based on the cloud top bright and the rainfall information.
5. A method of initializing a plateau vortex with balanced power and hydrothermal structure as claimed in claim 1, wherein: the step of plateau vortex initialization is to take a vortex three-dimensional hydro-thermal structure and a reconstructed vortex three-dimensional power structure as observation data, assimilate the observation data by using a three-dimensional variation assimilation technology after quality control, and enter a numerical forecasting mode, so that a new vortex with a coordinated initial moment dynamic field/structure and a hydro-thermal field/structure is formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311293641.6A CN117574748B (en) | 2023-10-08 | 2023-10-08 | Plateau vortex initialization method with balanced power and hydrothermal structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311293641.6A CN117574748B (en) | 2023-10-08 | 2023-10-08 | Plateau vortex initialization method with balanced power and hydrothermal structure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117574748A true CN117574748A (en) | 2024-02-20 |
| CN117574748B CN117574748B (en) | 2024-05-10 |
Family
ID=89888761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311293641.6A Active CN117574748B (en) | 2023-10-08 | 2023-10-08 | Plateau vortex initialization method with balanced power and hydrothermal structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574748B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106547840A (en) * | 2016-10-13 | 2017-03-29 | 国家卫星气象中心 | A kind of parsing of global three-dimensional atmospheric data and management method |
| CN109657314A (en) * | 2018-12-09 | 2019-04-19 | 中国人民解放军国防科技大学 | Typhoon vortex initialization method based on data cycle assimilation and mode cycle integration |
| CN113191299A (en) * | 2021-05-14 | 2021-07-30 | 中国科学院大气物理研究所 | Vortex identification method and device, storage medium and electronic equipment |
| CN114483485A (en) * | 2022-02-24 | 2022-05-13 | 兰州大学 | Method for improving wind speed prediction of Nudging wind power plant observation data |
| CN115455666A (en) * | 2022-08-23 | 2022-12-09 | 广东省气象台(南海海洋气象预报中心) | Method and device for realizing vortex removal and vortex movement in grid weather forecast |
-
2023
- 2023-10-08 CN CN202311293641.6A patent/CN117574748B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106547840A (en) * | 2016-10-13 | 2017-03-29 | 国家卫星气象中心 | A kind of parsing of global three-dimensional atmospheric data and management method |
| CN109657314A (en) * | 2018-12-09 | 2019-04-19 | 中国人民解放军国防科技大学 | Typhoon vortex initialization method based on data cycle assimilation and mode cycle integration |
| CN113191299A (en) * | 2021-05-14 | 2021-07-30 | 中国科学院大气物理研究所 | Vortex identification method and device, storage medium and electronic equipment |
| CN114483485A (en) * | 2022-02-24 | 2022-05-13 | 兰州大学 | Method for improving wind speed prediction of Nudging wind power plant observation data |
| CN115455666A (en) * | 2022-08-23 | 2022-12-09 | 广东省气象台(南海海洋气象预报中心) | Method and device for realizing vortex removal and vortex movement in grid weather forecast |
Non-Patent Citations (2)
| Title |
|---|
| ZHANG FM 等: "Genesis of Tibetan Plateau Vortex: Roles of Surface Diabatic and Atmospheric Condensational Latent Heating", JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY, vol. 58, no. 12, 31 December 2019 (2019-12-31), pages 2633 - 2651 * |
| 王溪雯 等: "青藏高原西部一次高原涡生成的数值模拟研究", 干旱气象, vol. 39, no. 1, 28 February 2021 (2021-02-28) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117574748B (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liston et al. | A meteorological distribution system for high-resolution terrestrial modeling (MicroMet) | |
| Kiladis et al. | Origin of the South Pacific convergence zone | |
| Sultan et al. | The West African monsoon dynamics. Part I: Documentation of intraseasonal variability | |
| Lélé et al. | Analysis of low-level atmospheric moisture transport associated with the West African monsoon | |
| Wu et al. | A numerical study of Hurricane Erin (2001). Part I: Model verification and storm evolution | |
| Chu et al. | Nonlinear, intraseasonal phases of the East Asian summer monsoon: Extraction and analysis using self-organizing maps | |
| Pan et al. | Characteristics and mechanism of the 10–20-day oscillation of spring rainfall over southern China | |
| Angevine et al. | Land surface spinup for episodic modeling | |
| Miller et al. | Topographic speed-up effects and observed roof damage on Bermuda following Hurricane Fabian (2003) | |
| Dale et al. | Atmospheric forcing of sea ice anomalies in the Ross Sea polynya region | |
| Yang et al. | Accelerating the EnKF spinup for typhoon assimilation and prediction | |
| Aoki et al. | Subsurface subtropical fronts of the North Pacific as inherent boundaries in the ventilated thermocline | |
| Byun et al. | Numerical simulations of heavy rainfall over central Korea on 21 September 2010 using the WRF model | |
| Ya-li et al. | Synoptic characteristics related to warm-sector torrential rainfall events in South China during the annually first rainy season | |
| T'u et al. | The advance and retreat of the summer monsoon in China | |
| CN117574748B (en) | Plateau vortex initialization method with balanced power and hydrothermal structure | |
| Park et al. | Accuracy evaluation of daily-gridded ASCAT satellite data around the Korean Peninsula | |
| Nakagawa | Outline of the high resolution global model at the Japan Meteorological Agency | |
| Kvamstø | An investigation of diagnostic relations between stratiform fractional cloud cover and other meteorological parameters in numerical weather prediction models | |
| Wijayanti et al. | The Impact of the Interaction between Madden-Julian Oscillation and Cold Surge, on Rainfall over Western Indonesia | |
| Paton et al. | A metropolitan wind resource assessment for Bangkok, Thailand part 1: Wind resource mapping | |
| Hou et al. | Climatological characteristics of frontogenesis and related circulations over East China in June and July | |
| CN115826089B (en) | East-west vortex boundary extraction method based on vortex tracking data | |
| CN113012285B (en) | Vortex initial field construction method for south China sea typhoon scale-division mixing | |
| Moujane et al. | Analysis of high spatial and temporal surface winds from Aladin model and from remotely sensed data over the Canarian upwelling region |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |