Method and device for producing reanalysis products of kilometer-scale land-air weak coupling in target area

Abstract

This invention discloses a method and apparatus for producing kilometer-scale land-atmosphere weakly coupled reanalysis products for a target area in the field of meteorology. The method includes constructing an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target area based on a kilometer-scale land-atmosphere weakly coupled numerical model system; employing hourly standard initialization of convective-scale atmospheric reanalysis and variational assimilation of assimilated observation datasets; and obtaining hourly kilometer-scale 3D atmospheric reanalysis products, soil temperature and soil moisture reanalysis products, and 1-hour precipitation forecast products for the target area through variational assimilation techniques, numerical forecast models, and land surface assimilation analysis techniques. This invention utilizes land-atmosphere weakly coupled reanalysis technology based on complex interactions at the atmosphere-land interface, enabling bidirectional feedback between the atmospheric and terrestrial spheres, accurately characterizing key aspects of the energy and water cycles, and improving the determinism of the target area reanalysis products in representing extreme events closely related to land surface processes.

Description

Technical Field

[0001] This invention relates to the field of meteorological technology, specifically to a method and apparatus for producing land-atmosphere weak coupling reanalysis products at the kilometer scale for a target area. Background Technology

[0002] Atmospheric reanalysis is a core technology for generating long-term, homogeneous, and high-precision climate benchmark datasets. Through a fixed data assimilation system and numerical models, it fuses historical observations with model forecasts to reproduce spatially complete and temporally continuous multivariate atmospheric conditions. This product is not only a key foundation for weather forecasting and research but is also widely used in environmental, agricultural, and energy fields. In recent years, large-scale AI weather forecasting models, such as FourCastNet, GraphCast, Pangu, and Fengwu, have all used atmospheric reanalysis products as training data, demonstrating forecast accuracy and speed advantages comparable to or even surpassing traditional numerical models.

[0003] Since the 1990s, the successive introduction of products such as the US NCEP / NCAR reanalysis, the European Centre for Reanalysis (ERA5), the Japanese JRA-55, and the Chinese CRA-40 / CMA-CRA has brought significant progress to atmospheric reanalysis. The spatiotemporal resolution of reanalysis products has continuously improved, with temporal resolution decreasing from 6-hour increments to 1-hour increments, and spatial resolution decreasing from 125 km to 13 km. Simultaneously, various countries and organizations have begun developing regional high-resolution atmospheric reanalysis to better describe local small-scale weather and climate change characteristics. Examples include the US North American Regional Reanalysis (NARR), the European Regional Reanalysis (HRRRE), the Japanese Regional Reanalysis (NHM-LETKF), the East Asian Regional Reanalysis (EARS) developed by the China Meteorological Administration's Institute of Meteorological Sciences based on WRF / GSI, and the 3km China Regional Atmospheric Reanalysis (CMA-RRAv1.0) developed by the domestic numerical model CMA-MESO for the China Meteorological Administration's Earth System Numerical Prediction. The quality and accuracy of these products have been continuously improving.

[0004] However, current regional reanalysis has inherent limitations in its technical paradigm: land surface conditions (such as soil moisture and temperature) are treated as simple boundary conditions in traditional atmospheric reanalysis. The accurate characterization of these key state variables directly determines the simulation accuracy of land-atmosphere energy and water exchange. Due to the spatiotemporal discontinuity of site observations and the inherent uncertainties in land surface models, the existing framework lacks continuous dynamic optimization of key variables such as soil moisture and temperature, making it difficult to accurately characterize the core aspects of the energy and water cycle. Therefore, developing technologies that can integrate observations and model simulations to achieve dynamic assimilation of land surface conditions is a key path to overcome the limitations of current reanalysis and improve the ability of land-atmosphere coupled simulation and extreme event prediction. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for producing land-atmosphere weakly coupled reanalysis products at the kilometer scale in a target area. By optimally fusing observational data and land surface process simulation, a land surface state field that is spatiotemporally continuous, physically consistent, and more accurate is generated. This optimizes the estimation of key land surface state variables such as soil moisture. By utilizing bidirectional coupling between land and air, the analysis of atmospheric variables and the initial field are improved. This land-atmosphere coupled reanalysis technology accurately describes the complex interactions between the atmosphere and land interface, breaking through current technical bottlenecks and meeting the demand for high-precision data.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for producing a land-atmosphere weakly coupled reanalysis product at the kilometer scale for a target area, the method comprising: Based on the kilometer-scale land-atmosphere weakly coupled numerical model system, the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weakly coupled reanalysis product for the target area were constructed, resulting in an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target area, including observation data preprocessing module, variational assimilation module, land surface assimilation module, and numerical prediction model module. Collect hourly atmospheric reanalysis products at the convective scale for the target area, hourly multi-source observation datasets for long-term series, weather radar network observations, and satellite observation datasets, and perform standard initialization on the hourly atmospheric reanalysis products at the convective scale. The variational assimilation module first uses standard initialization convective-scale atmospheric reanalysis and assimilable analysis observation datasets for variational assimilation; subsequently, it uses the target area's 1-hour future 3D atmospheric forecast field and assimilable analysis observation datasets output by the numerical weather prediction model module for variational assimilation, to obtain hourly kilometer-scale 3D atmospheric reanalysis products for the target area. The land surface assimilation analysis module for the first time uses standard initialized convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and surface 2m temperature and 2m humidity observations to obtain soil temperature and soil humidity reanalysis products for the target area; subsequently, it uses the soil temperature and soil humidity forecast products for the target area for the next 1 hour output by the numerical weather prediction model module, standard initialized convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, surface 2m temperature and 2m humidity observations, and the land surface assimilation module to obtain soil temperature and soil humidity reanalysis products. Based on the hourly kilometer-scale 3D atmospheric reanalysis product, soil temperature and soil moisture reanalysis product, standardized convection reanalysis product, and numerical model forecast module, the next 1-hour precipitation forecast product and soil temperature and soil moisture forecast product are obtained.

[0007] As a further aspect of the present invention, the range of the kilometer scale is less than 10 km.

[0008] As a further aspect of the present invention: the land-atmosphere weakly coupled numerical model system is the CMA_MESO system.

[0009] As a further aspect of the present invention, the numerical prediction model includes a land surface model.

[0010] As a further aspect of the present invention: the hourly atmospheric reanalysis product at the convective scale is an atmospheric reanalysis product for a 3km area.

[0011] As a further aspect of the present invention: the long-term time series hourly multi-source observation dataset includes radiosonde, aircraft reports, ship datasets, and ground observation datasets, wherein the ground observation dataset includes ground automatic station datasets.

[0012] As a further aspect of this invention: hourly atmospheric reanalysis products at the convective scale for the target area, hourly multi-source observation datasets over a long time series, weather radar network observations, and satellite observation data datasets are collected, and the hourly atmospheric reanalysis products at the convective scale are initialized using standard methods, including: The observation data preprocessing module performs data decoding, format conversion, element selection, and homogenization on the observation dataset, and identifies and removes non-meteorological information, identifies outliers, and controls the quality of sparse data to obtain an assimilation analysis observation dataset.

[0013] As a further aspect of the present invention: the hourly-kilometer-scale 3D atmospheric reanalysis product includes wind, temperature, pressure, and humidity reanalysis products.

[0014] Secondly, the present invention provides a device for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in a target area, the device comprising: The building unit is used to construct the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weak coupling reanalysis product for the target area based on the kilometer-scale land-atmosphere weak coupling numerical model system, and obtain the hourly kilometer-scale land-atmosphere weak coupling reanalysis system adapted to the target area. It includes an observation data preprocessing module, a variational assimilation module, a land surface assimilation module, and a numerical prediction model module. The collection unit is used to collect hourly atmospheric reanalysis products at the convective scale, hourly multi-source observation datasets of long-term series, weather radar network observations, and satellite observation data datasets of the target area, and to perform standard initialization on the hourly atmospheric reanalysis products at the convective scale. The variational assimilation module, for the first time, employs standard initialization of convective-scale atmospheric reanalysis and assimilable analysis observation datasets for variational assimilation. Subsequently, it uses the 3D atmospheric forecast field for the target area for the next hour output by the numerical weather prediction model module and assimilable analysis observation datasets for variational assimilation, obtaining hourly, kilometer-scale 3D atmospheric reanalysis products for the target area. The land surface assimilation analysis module, for the first time, uses standard initialization of convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and ground 2m temperature and 2m humidity observations to obtain soil temperature and soil moisture content for the target area. Humidity reanalysis products; subsequently, soil temperature and soil moisture reanalysis products are obtained using the target area's 1-hour soil temperature and soil moisture forecast products output by the numerical weather prediction model module, the standard initialized convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, the surface 2m temperature and 2m humidity observation and land surface assimilation modules; based on the hourly kilometer-scale 3D atmospheric reanalysis products, soil temperature and soil moisture reanalysis products, standardized convective reanalysis products, and the numerical weather prediction module, precipitation forecast products and soil temperature and soil moisture forecast products for the next 1 hour are obtained.

[0015] Thirdly, the present invention also provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; the memory storing instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described above.

[0016] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the methods described above.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention, based on a kilometer-scale land-atmosphere weakly coupled numerical model system, constructs an hourly kilometer-scale land-atmosphere weakly coupled reanalysis product development process and parameter configuration for the target area. This results in an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target area. Then, based on the constructed analysis system, standard initialization processing is performed on the hourly convective-scale atmospheric reanalysis product, long-term series hourly multi-source observation datasets, weather radar network observations, satellite observation datasets, and hourly convective-scale atmospheric reanalysis product to obtain a 3D atmospheric reanalysis product at the hourly kilometer scale for the target area. This achieves a two-way feedback process between the atmosphere and land spheres, accurately characterizing key aspects of the energy and water cycles, and improving the certainty in representing extreme events closely related to land surface processes. Attached Figure Description

[0018] Figure 1This diagram illustrates the steps of the method for producing the target area kilometer-scale land-atmosphere weak coupling reanalysis product of the present invention. Figure 2 This is a flowchart of the production device for the target area kilometer-scale land-atmosphere weak coupling reanalysis product of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example: Please see Figure 1 In this embodiment of the invention, a method for producing a land-atmosphere weak coupling reanalysis product at the kilometer scale for a target area includes the following steps: S1: Based on the kilometer-scale land-atmosphere weakly coupled numerical model system, the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weakly coupled reanalysis product for the target area are constructed, resulting in an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target area, including observation data preprocessing module, variational assimilation module, land surface assimilation module, and numerical prediction model module.

[0021] The target region can be China, and the land-atmosphere weakly coupled numerical model system is the CMA_MESO system independently developed by the Earth System Numerical Prediction Center of the China Meteorological Administration.

[0022] S2: Collect hourly atmospheric reanalysis products at the convective scale for the target area, hourly multi-source observation datasets for long-term series, weather radar network observations, and satellite observation datasets, and perform standard initialization on the hourly atmospheric reanalysis products at the convective scale.

[0023] The horizontal resolution at the convective scale is 3km-10km. The standard initialization module for the convective scale reanalysis product decodes the hourly atmospheric reanalysis product at the convective scale in GRIB2 format and interpolates the decoded data to the numerical model grid points of the target area at the kilometer scale to achieve standard initialization. The observation data preprocessing module performs data decoding, format conversion, element selection, and homogenization on the observation dataset. At the same time, it carries out data quality control such as non-meteorological information identification and removal, outlier identification, and sparsification to obtain the assimilation analysis observation dataset.

[0024] S3: The variational assimilation module uses standard initialization of convective-scale atmospheric reanalysis and assimilable analysis observation datasets for variational assimilation for the first time; subsequently, the target area's 1-hour future 3D atmospheric forecast field and assimilable analysis observation datasets output by the numerical weather prediction model module are used for variational assimilation to obtain hourly kilometer-scale 3D atmospheric reanalysis products for the target area.

[0025] S4: The land surface assimilation analysis module for the first time uses standard initialization convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and surface 2m temperature and 2m humidity observations to obtain soil temperature and soil humidity reanalysis products for the target area; subsequently, it uses the soil temperature and soil humidity forecast products for the target area for the next 1 hour output by the numerical weather prediction model module, standard initialization convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, surface 2m temperature and 2m humidity observations, and the land surface assimilation module to obtain soil temperature and soil humidity reanalysis products.

[0026] S5: Based on hourly kilometer-scale 3D atmospheric reanalysis products, soil temperature and soil moisture reanalysis products, standardized convection reanalysis products, and numerical model forecasting modules, it obtains 1-hour precipitation forecast products, soil temperature and soil moisture forecast products.

[0027] In this embodiment, the kilometer scale is 0~10km, the hourly atmospheric reanalysis product at the convection scale is the atmospheric reanalysis product of the 3km area, and the hourly kilometer-scale 3D atmospheric reanalysis product includes wind, temperature, pressure, and humidity reanalysis products.

[0028] In this embodiment, step S2 includes: the observation data preprocessing module performs data decoding, format conversion, element selection, and homogenization on the observation dataset, and identifies and removes non-meteorological information, identifies outliers, and controls the quality of sparse data to obtain the assimilation analysis observation dataset.

[0029] In some embodiments, the variational assimilation module derives the objective function based on Bayesian posterior probability, and then uses the built-in observation data assimilation operator to achieve variational assimilation of different types and spatiotemporal resolutions of observation data in the assimilable analysis observation dataset. Based on this, the 3D atmospheric forecast field of the target area for the next hour output by the forecast module is adjusted and optimized to obtain the hourly convective-scale 3D atmospheric reanalysis product (such as wind, temperature, pressure, humidity, etc.) of the target area.

[0030] For example, the above objective function can be expressed as: In the formula, For the analysis field Relative to background field The increment; , as background scene and observation field The distance between them; For observation data assimilation operators; and These are the background error covariance matrix and the observation error covariance matrix, respectively; (•) -1 and(•) T These represent the operations of matrix inversion and transpose, respectively. and These are two weak constraints that make it easier to perform small- to medium-scale analysis.

[0031] In some embodiments, the land surface assimilation module employs the optimal interpolation (OI) assimilation method, which seeks the optimal estimate of the analysis field in a statistical sense. The land surface soil moisture assimilation analysis is independent for each soil layer, and the increase in soil moisture content for each layer is obtained according to the following formula: In the formula, It is the moisture content of the i-th soil layer. and These are the analytical coefficients of the i-th soil layer, reflecting the temperature at 2m ( ) and 2m humidity ( The contribution of the incremental change in soil moisture content to the increase in each soil layer is calculated using the following formula: in, represent and The correlation between forecast errors and These are the standard deviations of the forecast and analysis errors, respectively. and These are empirical coefficients used to reduce analytical coefficients when atmospheric errors are not caused by soil errors. This is used to simulate the weak correlation between changes in soil moisture content and the atmosphere under nighttime and winter conditions, using the following formula: in, It analyzes the average solar zenith angle over the first 6 hours. It is a constant. At the same time, under conditions of weak cloud cover and radiation, the coefficient of OI will also decrease. Therefore, Douville used atmospheric transmittance... and the average downward shortwave radiation from the ground surface in the 6 hours prior to analysis Add to OI coefficients The empirical coefficient is calculated using the following formula: In the formula, It is the solar constant. and Both are constants, 0.2 and 0.9 respectively.

[0032] In the OI method, the temperature analysis of topsoil and subsoil is only related to the air temperature at two meters, and is calculated using the empirical coefficient F1, as shown in the following formula: In the formula, It is the soil temperature assimilation coefficient. It is the temperature of the i-th soil layer. The temperature is 2 meters.

[0033] In some embodiments, the numerical weather prediction model (including the land surface model) module can process the 3D atmospheric reanalysis product output by the variational assimilation module and the soil temperature and soil moisture reanalysis product output by the land surface assimilation module based on the control equations that take into account the shallow atmospheric approximation, moist air, and topographical following coordinates, calculate the 3D atmospheric forecast field for the target area for the next hour, and generate the precipitation, soil temperature, and soil moisture forecast products for the target area for the next hour through physical parameterization processes and the land surface model.

[0034] For example, the governing equations for the above-mentioned shallow atmospheric approximation, moist air, and altitude topography following coordinates are expressed as follows: In the formula, , , These are velocity vectors in the east-west, north-south, and vertical directions, respectively. It is a dimensionless air pressure; Potential temperature; Temperature is the virtual potential. , These represent the longitude and latitude in spherical coordinates, respectively. The distance between the point mass and the Earth's center; For Coriolis force parameters, Here are the parameters for the latitudinal Coriolis force; g is the acceleration due to gravity; Cp is the specific heat capacity at constant pressure. Switch for curvature correction term; For the Coriolis force switch; Static balancing switch; , , Frictional forces in the east-west, north-south, and vertical directions; For specific heat ratio, and R is the gas constant; For three-dimensional divergence; This is a correction to a flux or source / sink term related to potential temperature, and , As a heat source and sink; The terrain height follows the coordinate height; The terrain height follows the vertical velocity in the coordinate system; This represents the Earth's radius after applying the thin-layer approximation. and These are the terrain height and the top of the model layer, respectively. The height of the top of the model layer above the ground. The model layer top is the coordinate height that follows the terrain. , It is the slope of the terrain. For time step.

[0035] At the same time, the forecasting module also uses water vapor or other condensates ( The equation can be expressed as: In the formula, This refers to water vapor or other water condensates; the meanings of other parameters are the same as above.

[0036] For example, the land surface model uses the Noah land surface model.

[0037] In the Noah land surface model, the soil is divided into four layers, with thicknesses of 0.1, 0.3, 0.6, and 1.0 meters from the top to the bottom, respectively. Land surface heat flux is calculated using the soil temperature heat diffusion formula, as follows: in, It is thermal conductivity. It is the soil volumetric water content. It is volumetric heat capacity. It's the soil temperature. It is time. It refers to soil depth.

[0038] Noah uses the following formula to calculate soil volumetric water content: in, It is the soil water diffusivity. It is the water conductivity. It represents the sources and sinks of soil water (precipitation, evaporation, etc.). The formula for determining the soil moisture content of each layer can be obtained from the above formula as follows: In the formula, It is the thickness of the i-th soil layer. It's precipitation. It is surface outflow. It is the interception of plants. It is the evaporation of the i-th layer of soil.

[0039] To facilitate further understanding, the above method will be explained in detail below using the target region of China as an example: (1) Obtain a weakly coupled land-atmosphere numerical model system at the kilometer scale (<10km) (The weakly coupled land-atmosphere numerical model system used in this patent is the CMA_MESO system independently developed by the Earth System Numerical Prediction Center of China Meteorological Administration). (2) Based on the kilometer-scale land-atmosphere weak coupling numerical model system, the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weak coupling reanalysis product for the target area are constructed, and the hourly kilometer-scale land-atmosphere weak coupling reanalysis system adapted to the target area is obtained, including the observation data preprocessing module, variational assimilation module, land surface assimilation module, and numerical prediction model (including land surface model) module. (3) Collect hourly atmospheric reanalysis products at the convective scale of the target area (the hourly atmospheric reanalysis products at the convective scale used in this patent are the 3km regional atmospheric reanalysis products independently developed by the Earth System Numerical Prediction Center of the China Meteorological Administration), hourly multi-source observation datasets of long-term series (such as radiosonde, aircraft reports, ship data, ground (including automatic ground stations) observation datasets, weather radar network observations, satellite observation data datasets, etc.), and perform standard initialization on the hourly atmospheric reanalysis products at the convective scale.

[0040] (4) The observation data preprocessing module performs data decoding, format conversion, element selection, and homogenization on the observation dataset. At the same time, it carries out data quality control such as non-meteorological information identification and removal, outlier identification, and sparsification to obtain the assimilation analysis observation dataset. (5) The variational assimilation module first uses standard initialization convective-scale atmospheric reanalysis and assimilable analysis observation dataset for variational assimilation. Then, it uses the target area's 3D atmospheric forecast field for the next hour output by the numerical forecast model (including the land surface model) module and the assimilable analysis observation dataset for variational assimilation to obtain the hourly kilometer-scale 3D atmospheric reanalysis product of the target area. (6) The land surface assimilation analysis module first used standard initialization convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and ground 2m temperature and 2m humidity observation to obtain soil temperature and soil humidity reanalysis products for the target area; subsequently, the soil temperature and soil humidity forecast products for the next hour of the target area output by the numerical weather prediction model (including the land surface model) module, the standard initialization convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, ground 2m temperature and 2m humidity observation and the land surface assimilation module were used to obtain soil temperature and soil humidity reanalysis products.

[0041] (7) Based on the above hourly kilometer-scale 3D atmospheric reanalysis products (wind, temperature, pressure, humidity, etc.), soil temperature and soil moisture reanalysis products, standardized convection reanalysis products, and numerical model forecasting modules (including land surface models), obtain the precipitation forecast products for the next 1 hour, and the soil temperature and soil moisture forecast products.

[0042] It is understandable that the specific details of steps (1) to (7) here are similar to those above, and will not be repeated here for the sake of brevity.

[0043] Based on this embodiment, it can be seen that the present invention achieves at least the following technical effects: This invention, based on a kilometer-scale land-atmosphere weakly coupled numerical model system, constructs an hourly kilometer-scale land-atmosphere weakly coupled reanalysis product development process and parameter configuration for the target region. This results in an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target region. Then, based on the constructed analysis system, standard initialization processing is performed on the hourly convective-scale atmospheric reanalysis product, long-term series hourly multi-source observation datasets, weather radar network observations, satellite observation datasets, and hourly convective-scale atmospheric reanalysis product to obtain a 3D atmospheric reanalysis product at the hourly kilometer scale for the target region. This achieves a two-way feedback process between the atmosphere and land spheres, accurately characterizing key aspects of the energy and water cycles, improving the certainty in representing extreme events closely related to land surface processes, and providing solid data support for meteorological research, disaster prevention and mitigation, and climate change response in my country, thus promoting the high-quality development of my country's meteorological undertakings. Meanwhile, the high-precision, high-spatiotemporal-resolution reanalysis products generated by the production device provided in this embodiment can also provide scientific basis for multiple industries such as agriculture, water conservancy, transportation, and energy, helping various sectors of the country to make more accurate and effective decisions in response to complex climate and environmental changes, further enhancing the country's comprehensive disaster prevention and mitigation capabilities, and ensuring the sustainable development of the country's economy and society.

[0044] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0045] The above is an introduction to the method embodiments. The following describes the present invention further through device embodiments.

[0046] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for producing land-atmosphere weakly coupled reanalysis products at the kilometer scale for a target area, characterized in that, The method includes: Based on the kilometer-scale land-atmosphere weakly coupled numerical model system, the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weakly coupled reanalysis product for the target area were constructed, resulting in an hourly kilometer-scale land-atmosphere weakly coupled reanalysis system adapted to the target area, including observation data preprocessing module, variational assimilation module, land surface assimilation module, and numerical prediction model module. Collect hourly atmospheric reanalysis products at the convective scale for the target area, hourly multi-source observation datasets for long-term series, weather radar network observations, and satellite observation datasets, and perform standard initialization on the hourly atmospheric reanalysis products at the convective scale. The variational assimilation module first uses standard initialization convective-scale atmospheric reanalysis and assimilable analysis observation datasets for variational assimilation; subsequently, it uses the target area's 1-hour future 3D atmospheric forecast field and assimilable analysis observation datasets output by the numerical weather prediction model module for variational assimilation, to obtain hourly kilometer-scale 3D atmospheric reanalysis products for the target area. The land surface assimilation analysis module for the first time uses standard initialized convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and surface 2m temperature and 2m humidity observations to obtain soil temperature and soil humidity reanalysis products for the target area; subsequently, it uses the soil temperature and soil humidity forecast products for the target area for the next 1 hour output by the numerical weather prediction model module, standard initialized convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, surface 2m temperature and 2m humidity observations, and the land surface assimilation module to obtain soil temperature and soil humidity reanalysis products. Based on the hourly kilometer-scale 3D atmospheric reanalysis product, soil temperature and soil moisture reanalysis product, standardized convection reanalysis product, and numerical model forecast module, the next 1-hour precipitation forecast product and soil temperature and soil moisture forecast product are obtained.

2. The method for producing land-atmosphere coupled reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The range of the kilometer scale is less than 10 km.

3. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The land-atmosphere weakly coupled numerical model system is the CMA_MESO system.

4. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The numerical prediction model includes a land surface model.

5. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The hourly atmospheric reanalysis product at the convective scale is an atmospheric reanalysis product for a 3km area.

6. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The long-term hourly multi-source observation dataset includes radiosonde, aircraft reports, ship datasets, and ground observation datasets, with the ground observation dataset including automatic ground station datasets.

7. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, Collect hourly atmospheric reanalysis products at the convective scale for the target area, hourly multi-source observation datasets over long time series, weather radar network observations, and satellite observation datasets, and perform standard initialization on the hourly convective-scale atmospheric reanalysis products, including: The observation data preprocessing module performs data decoding, format conversion, element selection, and homogenization on the observation dataset, and identifies and removes non-meteorological information, identifies outliers, and controls the quality of sparse data to obtain an assimilation analysis observation dataset.

8. The method for producing land-atmosphere weak coupling reanalysis products at the kilometer scale in the target area according to claim 1, characterized in that, The hourly-kilometer-scale 3D atmospheric reanalysis product includes wind, temperature, pressure, and humidity reanalysis products.

9. A device for producing land-atmosphere weakly coupled reanalysis products at the kilometer scale in a target area, characterized in that, The device includes: The building unit is used to construct the development process and parameter configuration of the hourly kilometer-scale land-atmosphere weak coupling reanalysis product for the target area based on the kilometer-scale land-atmosphere weak coupling numerical model system, and obtain the hourly kilometer-scale land-atmosphere weak coupling reanalysis system adapted to the target area. It includes an observation data preprocessing module, a variational assimilation module, a land surface assimilation module, and a numerical prediction model module. The collection unit is used to collect hourly atmospheric reanalysis products at the convective scale, hourly multi-source observation datasets of long-term series, weather radar network observations, and satellite observation data datasets of the target area, and to perform standard initialization on the hourly atmospheric reanalysis products at the convective scale. The variational assimilation module, for the first time, employs standard initialization of convective-scale atmospheric reanalysis and assimilable analysis observation datasets for variational assimilation. Subsequently, it uses the 3D atmospheric forecast field for the target area for the next hour output by the numerical weather prediction model module and assimilable analysis observation datasets for variational assimilation, obtaining hourly, kilometer-scale 3D atmospheric reanalysis products for the target area. The land surface assimilation analysis module, for the first time, uses standard initialization of convective-scale atmospheric reanalysis 2m temperature, 2m humidity, soil temperature and soil humidity products, and ground 2m temperature and 2m humidity observations to obtain soil temperature and soil moisture content for the target area. Humidity reanalysis products; subsequently, soil temperature and soil moisture reanalysis products are obtained using the target area's 1-hour soil temperature and soil moisture forecast products output by the numerical weather prediction model module, the standard initialized convective-scale atmospheric reanalysis 2m temperature and 2m humidity products, the surface 2m temperature and 2m humidity observation and land surface assimilation modules; based on the hourly kilometer-scale 3D atmospheric reanalysis products, soil temperature and soil moisture reanalysis products, standardized convective reanalysis products, and the numerical weather prediction module, precipitation forecast products and soil temperature and soil moisture forecast products for the next 1 hour are obtained.

Images