CN117933095A - Earth surface emissivity real-time inversion and assimilation method based on machine learning - Google Patents
Earth surface emissivity real-time inversion and assimilation method based on machine learning Download PDFInfo
- Publication number
- CN117933095A CN117933095A CN202410319958.0A CN202410319958A CN117933095A CN 117933095 A CN117933095 A CN 117933095A CN 202410319958 A CN202410319958 A CN 202410319958A CN 117933095 A CN117933095 A CN 117933095A
- Authority
- CN
- China
- Prior art keywords
- earth surface
- assimilation
- emissivity
- satellite
- surface emissivity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000010801 machine learning Methods 0.000 title claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 27
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000004088 simulation Methods 0.000 claims abstract description 3
- 238000004364 calculation method Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 238000002834 transmittance Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 7
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a machine learning-based earth surface emissivity real-time inversion and assimilation method, which comprises the following steps: calculating the corresponding satellite simulation radiance when no cloud distribution exists in the satellite observation view field; training to obtain an earth surface emissivity inversion model; taking the earth surface temperature at the current assimilation moment and the actual satellite observation emissivity as inputs of an earth surface emissivity inversion model, providing the earth surface emissivity output by the earth surface emissivity inversion model for a satellite assimilation module in an assimilation system, and inverting the earth surface emissivity in satellite assimilation; and carrying out full earth surface assimilation on the selected satellite observation data to obtain a numerical mode analysis field at the current assimilation moment for deterministic prediction. The invention can introduce real-time earth surface emissivity information in the assimilation process, realize more full utilization of earth surface sensitive satellite data, effectively improve the accuracy of earth surface emissivity when assimilating the satellite data, and realize the full earth surface assimilation of the satellite data in an assimilation system.
Description
Technical Field
The invention belongs to the technical field of atmospheric science, and particularly relates to a machine learning-based earth surface emissivity real-time inversion and assimilation method.
Background
With the rapid development of computer technology, the continuous improvement of numerical weather forecast models and the continuous improvement of data assimilation technologies, the numerical forecast model has become an important means for providing future weather forecast results for various business forecast centers worldwide at present. Improving the accuracy of the numerical mode initial field through data assimilation is a key to improving the accuracy of numerical mode forecasting results, wherein satellite data is one of the observations with the greatest contribution to improving numerical mode forecasting level. The satellite data is acquired by a weather remote sensing instrument carried on a static satellite and a polar orbit satellite, and contains weather related information which is extremely rich from the ground surface to the upper air of the atmosphere. At present, the utilization rate of satellite data at home and abroad is only about 10%, and a large amount of satellite data cannot be used for assimilation, so that the assimilation of satellite data is developing to all sky and all earth surfaces, the satellite data is fully utilized, and the numerical mode forecasting level is further improved. Full sky assimilation has been a lot of research results, and many business centers have realized full sky assimilation of satellite data in business, and full earth's surface assimilation is still under development at present.
One of the major difficulties in total surface assimilation is the inaccuracy of surface information. The earth surface information such as earth surface emissivity has important influence on satellite data assimilation, but the approach of accurately observing the earth surface information in real time is lacking at present, and the numerical mode of accurately forecasting the change trend of the earth surface information is also lacking, so that satellite observation sensitive to earth surface cannot be fully assimilated. At present, two main methods for processing the surface emissivity in data assimilation exist, namely, the surface emissivity is calculated through a climatic state surface emissivity data set, and the data set is a long-term average emissivity situation analyzed through historical data, but the change of the surface emissivity can not be fully reflected when the surface emissivity changes; secondly, emissivity information is calculated in real time through a numerical mode, but the method can only be applied to areas with low emissivity calculation complexity such as ocean surfaces, and for land areas with obvious surface feature space-time variation and extremely high calculation complexity, the available emissivity mode is lacking.
Disclosure of Invention
The technical problems to be solved are as follows: aiming at the technical problem that the accurate emissivity information of the whole earth surface in each assimilation system is difficult to obtain. The invention discloses a real-time earth surface emissivity inversion and assimilation method based on machine learning, which can introduce real-time earth surface emissivity information in the assimilation process, realize more full utilization of earth surface sensitive satellite data, effectively improve the accuracy of earth surface emissivity when assimilating the satellite data, and realize full earth surface assimilation of the satellite data in an assimilation system.
The technical scheme is as follows:
the invention discloses a machine learning-based earth surface emissivity real-time inversion and assimilation method, which comprises the following steps:
S1, combining the earth surface emissivity data set and the ERA5 re-analyzing the earth surface temperature and the atmospheric state variable in the data set, and calculating the corresponding satellite simulated emissivity when no cloud distribution exists in the satellite observation view field by using the following formula ：
In the method, in the process of the invention,Is the surface emissivity,/>The channel detection frequency corresponding to the satellite simulated emissivity L is T is the atmospheric temperature from the earth surface to the satellite, and the frequency is/is equal to the atmospheric temperature of the satelliteFor atmospheric transmittance from the earth's surface to the satellite,/>For the atmospheric transmittance from the atmosphere of each layer to the satellite,/>B is planck function for surface temperature;
S2, using the earth surface temperature in the ERA5 analysis data set and the satellite simulated radiance data calculated in the step S1 as an earth surface emissivity inversion method training set based on machine learning, setting the number of hidden layers according to the actual earth surface type, training by using the earth surface emissivity data set as a true value, and training a machine learning training model to obtain a trained earth surface emissivity inversion model; building a trained earth surface emissivity inversion model into an assimilation system;
S3, obtaining the earth surface temperature at the current assimilation time from the numerical mode background field at the current assimilation time, taking the earth surface temperature at the current assimilation time and the actual satellite observation emissivity as the input of an earth surface emissivity inversion model, providing the earth surface emissivity output by the earth surface emissivity inversion model for a satellite assimilation module in an assimilation system, and inverting the earth surface emissivity in satellite assimilation;
S4, carrying out full earth surface assimilation on the selected satellite observation data based on earth surface emissivity information obtained by inversion in the assimilation system, and obtaining a numerical mode analysis field at the current assimilation moment;
s5, carrying out deterministic prediction by using the numerical mode analysis field obtained in the step S4, generating a numerical mode background field at the next assimilation moment, returning to the step S3, and executing the full earth surface assimilation of the satellite observation data at the next assimilation moment.
Further, in step S1, a USGS surface emissivity dataset is used.
Further, in step S2, in the training process, RMSE between the training result and the true value is calculated as accuracy to perform verification, and the machine learning parameters are adjusted according to the RMSE between the training result and the true value until the RMSE between the training result and the true value meets the prediction accuracy requirement.
Further, in step S2, the process of obtaining the trained earth surface emissivity inversion model includes:
performing inverse operation on a calculation formula of satellite simulation emissivity to obtain a calculation formula of earth surface emissivity, wherein the calculation formula is as follows:
；
The simplified process is as follows:
wherein f is the simulated emissivity by satellite Calculating the surface emissivity/>Is a nonlinear function of satellite simulated emissivity/>With the surface temperature/>And atmospheric transmittance/>Related to;
Establishing satellite simulated radiance based on machine learning training model And earth emissivity/>Relationship between the acquisition function/>While limiting the surface emissivity/>, in the machine learning process by the following physical constraintsIs reasonable in the result:
。
The beneficial effects are that:
According to the earth surface emissivity real-time inversion and assimilation method based on machine learning, earth surface emissivity data matched with satellite observation is rapidly calculated in real time through satellite emissivity observation data and numerical mode earth surface temperature information at corresponding time. The earth surface emissivity inversion method with the new structure can be used as a module of an assimilation system, so that the real-time high-precision calculation requirement on earth surface emissivity in the whole earth surface assimilation of satellite data is met, and the satellite data is fully utilized in the assimilation system.
Drawings
FIG. 1 is a flow chart of a machine learning-based method for real-time inversion and assimilation of surface emissivity.
Detailed Description
The following examples will provide those skilled in the art with a more complete understanding of the invention, but are not intended to limit the invention in any way.
The invention discloses a machine learning-based earth surface emissivity real-time inversion and assimilation method, which comprises the following steps:
S1, combining the earth surface emissivity data set and the ERA5 re-analyzing the earth surface temperature and the atmospheric state variable in the data set, and calculating the corresponding satellite simulated emissivity when no cloud distribution exists in the satellite observation view field by using the following formula ：
In the method, in the process of the invention,Is the surface emissivity,/>The channel detection frequency corresponding to the satellite simulated emissivity L is T is the atmospheric temperature from the earth surface to the satellite, and the frequency is/is equal to the atmospheric temperature of the satelliteFor atmospheric transmittance from the earth's surface to the satellite,/>For the atmospheric transmittance from the atmosphere of each layer to the satellite,/>B is planck function for surface temperature;
S2, using the earth surface temperature in the ERA5 analysis data set and the satellite simulated radiance data calculated in the step S1 as an earth surface emissivity inversion method training set based on machine learning, setting the number of hidden layers according to the actual earth surface type, training by using the earth surface emissivity data set as a true value, and training a machine learning training model to obtain a trained earth surface emissivity inversion model; building a trained earth surface emissivity inversion model into an assimilation system;
S3, obtaining the earth surface temperature at the current assimilation time from the numerical mode background field at the current assimilation time, taking the earth surface temperature at the current assimilation time and the actual satellite observation emissivity as the input of an earth surface emissivity inversion model, providing the earth surface emissivity output by the earth surface emissivity inversion model for a satellite assimilation module in an assimilation system, and inverting the earth surface emissivity in satellite assimilation;
S4, carrying out full earth surface assimilation on the selected satellite observation data based on earth surface emissivity information obtained by inversion in the assimilation system, and obtaining a numerical mode analysis field at the current assimilation moment;
s5, carrying out deterministic prediction by using the numerical mode analysis field obtained in the step S4, generating a numerical mode background field at the next assimilation moment, returning to the step S3, and executing the full earth surface assimilation of the satellite observation data at the next assimilation moment.
The satellite data can detect various meteorological information from the earth surface to the satellite, and more detection channels sensitive to the earth surface are used for detecting earth surface information and information near the earth surface. When clouds appear under the field of view observed by the satellite, the clouds can absorb and scatter the radiation from the ground surface, so that the satellite cannot fully receive information such as emissivity of the ground surface. Therefore, the invention firstly screens the observation capable of detecting the earth surface information through cloud detection, and the cloud detection data is derived from cloud products in satellite materials：
。
When there is no cloud distribution in the field of view observed by the satellite, the earth surface information and the atmospheric information between the earth surface and the satellite are detected by the satellite and finally converted into the radiance information, and the satellite simulated radiance L can be calculated by the following formula:
In the method, in the process of the invention, Is the surface emissivity,/>Channel detection frequency corresponding to satellite analog emissivity L,/>For atmospheric transmittance from the earth's surface to the satellite,/>For the atmospheric transmittance from the atmosphere of each layer to the satellite,/>Is the surface temperature, and B is the planck function. In an assimilation system, the atmospheric state variables calculate the corresponding atmospheric transmittance through the corresponding modules in the observation operators, while/>And/>As a direct input in the observer. Wherein/>Can be obtained in real time by land mode, earth surface observation and other modes, and earth surface emissivity/>The corresponding acquisition path is lacking.
Based on the problem, the invention provides a method for acquiring the surface emissivity. Specifically:
the earth surface emissivity is obtained by performing inverse operation on a calculation formula of satellite simulated emissivity Is calculated according to the formula:
；
to transmit the surface emissivity Is simplified from the calculation formula of (a):
Wherein f is the simulated emissivity through satellite Calculating the surface emissivity/>Function of/>And the earth surface temperature/>And atmospheric transmittance/>Concerning, surface temperature/>Can be obtained by observing or numerical mode background fields,/>Directly related to the atmospheric conditions and can be calculated by means of the radiation transmission modes that are currently mature. Considering that the primitive function corresponding to the function f involves complex integral calculation, f is a highly complex nonlinear function and is difficult to directly solve by traditional mathematical and physical methods.
The machine learning method has unique advantages in dealing with highly nonlinear processes, in terms of surface emissivityThe invention proposes to use machine learning to help obtain the function f, build/>, to take advantage of the same calculationAnd/>Relationship between, while limiting machine learning/>, by physical constraintsRationality of results:
。
After inversion of the earth surface emissivity is completed, a trained earth surface emissivity inversion model is added into an assimilation system, and satellite observation is read in real time in the assimilation process Information for rapidly calculating corresponding earth surface emissivity/>Thereby providing support for earth surface sensitive satellite observation assimilation.
Referring to fig. 1, the present embodiment discloses one example of a machine learning-based earth surface emissivity inversion method, which can accurately calculate the earth surface emissivity required for assimilation of all-earth surface satellite materials in real time with high efficiency during assimilation. Specifically, the earth surface emissivity inversion method comprises the following steps:
(1) And calculating the corresponding satellite simulated radiance by using a satellite simulated radiance calculation formula by using a surface emissivity data set such as USGS and surface temperature and atmospheric state variables in an ERA5 analysis data set.
(2) And (2) using the surface temperature in the ERA5 analysis data set and the simulated emissivity data calculated in the step (1) as a surface emissivity inversion method training set based on machine learning, setting a proper hidden layer number and the like according to the actual surface type, training by using the actual surface emissivity data set such as USGS and the like as a true value, and calculating the RMSE between the training result and the true value as accuracy to test.
(3) And (3) adjusting machine learning parameters according to training results, and repeating the step (2) until parameters and settings meeting the precision requirements are obtained, and outputting a trained earth surface emissivity inversion model.
(4) The trained earth surface emissivity inversion model is built into an assimilation system, the earth surface temperature and the actual satellite observation emissivity in the numerical mode forecasting result are used as the input of the earth surface emissivity inversion model, the earth surface emissivity at the current assimilation time output by the earth surface emissivity inversion model is provided for a satellite assimilation module in the assimilation system, and the earth surface emissivity in satellite assimilation is inverted in real time and high precision.
(5) And (3) selecting proper satellite observation data based on the earth surface emissivity information provided in the assimilation system in the step (4), realizing the full earth surface assimilation of the satellite data, and obtaining a better numerical mode analysis field.
(6) And (3) carrying out deterministic prediction by using the numerical mode analysis field obtained in the step (5) to generate a numerical mode background field at the next assimilation moment so as to carry out satellite data full-surface assimilation at the next moment.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above examples, and all technical solutions belonging to the concept of the present invention belong to the protection scope of the present invention. It should be noted that modifications and adaptations to the invention without departing from the principles thereof are intended to be within the scope of the invention as set forth in the following claims.
Claims (4)
1. The earth surface emissivity real-time inversion and assimilation method based on machine learning is characterized by comprising the following steps of:
S1, combining the earth surface emissivity data set and the ERA5 re-analyzing the earth surface temperature and the atmospheric state variable in the data set, and calculating the corresponding satellite simulated emissivity when no cloud distribution exists in the satellite observation view field by using the following formula ：
In the method, in the process of the invention,Is the surface emissivity,/>The channel detection frequency corresponding to the satellite simulated emissivity L is T is the atmospheric temperature from the earth surface to the satellite, and the frequency is/is equal to the atmospheric temperature of the satelliteFor atmospheric transmittance from the earth's surface to the satellite,/>For the atmospheric transmittance from the atmosphere of each layer to the satellite,B is planck function for surface temperature;
S2, using the earth surface temperature in the ERA5 analysis data set and the satellite simulated radiance data calculated in the step S1 as an earth surface emissivity inversion method training set based on machine learning, setting the number of hidden layers according to the actual earth surface type, training by using the earth surface emissivity data set as a true value, and training a machine learning training model to obtain a trained earth surface emissivity inversion model; building a trained earth surface emissivity inversion model into an assimilation system;
S3, obtaining the earth surface temperature at the current assimilation time from the numerical mode background field at the current assimilation time, taking the earth surface temperature at the current assimilation time and the actual satellite observation emissivity as the input of an earth surface emissivity inversion model, providing the earth surface emissivity output by the earth surface emissivity inversion model for a satellite assimilation module in an assimilation system, and inverting the earth surface emissivity in satellite assimilation;
S4, carrying out full earth surface assimilation on the selected satellite observation data based on earth surface emissivity information obtained by inversion in the assimilation system, and obtaining a numerical mode analysis field at the current assimilation moment;
s5, carrying out deterministic prediction by using the numerical mode analysis field obtained in the step S4, generating a numerical mode background field at the next assimilation moment, returning to the step S3, and executing the full earth surface assimilation of the satellite observation data at the next assimilation moment.
2. The machine learning based surface emissivity real time inversion and assimilation method of claim 1, wherein in step S1, USGS surface emissivity dataset is used.
3. The machine learning-based earth surface emissivity real-time inversion and assimilation method according to claim 1, wherein in step S2, in the training process, RMSE between the training result and the true value is calculated as accuracy to check, and machine learning parameters are adjusted according to the RMSE between the training result and the true value until the RMSE between the training result and the true value meets the prediction accuracy requirement.
4. The machine learning-based real-time inversion and assimilation method for surface emissivity of claim 1, wherein in step S2, the process of obtaining a trained inversion model for surface emissivity comprises:
performing inverse operation on a calculation formula of satellite simulation emissivity to obtain a calculation formula of earth surface emissivity, wherein the calculation formula is as follows:
；
The simplified process is as follows:
wherein f is the simulated emissivity by satellite Calculating the surface emissivity/>Is a nonlinear function of satellite analog emissivityWith the surface temperature/>And atmospheric transmittance/>Related to;
Establishing satellite simulated radiance based on machine learning training model And earth emissivity/>Relationship between the acquisition function/>While limiting the surface emissivity/>, in the machine learning process by the following physical constraintsIs reasonable in the result:
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410319958.0A CN117933095B (en) | 2024-03-20 | 2024-03-20 | Earth surface emissivity real-time inversion and assimilation method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410319958.0A CN117933095B (en) | 2024-03-20 | 2024-03-20 | Earth surface emissivity real-time inversion and assimilation method based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117933095A true CN117933095A (en) | 2024-04-26 |
| CN117933095B CN117933095B (en) | 2024-06-18 |
Family
ID=90754129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410319958.0A Active CN117933095B (en) | 2024-03-20 | 2024-03-20 | Earth surface emissivity real-time inversion and assimilation method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117933095B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115046894A (en) * | 2022-05-18 | 2022-09-13 | 中国地质大学（武汉） | Cloud haze detection method based on radiation transmission simulation coupling machine learning technology |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101295022A (en) * | 2008-06-25 | 2008-10-29 | 中国农业科学院农业资源与农业区划研究所 | Method for ground surface temperature and emissivity inversion by remote sensing data ASTER |
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
| CN105930664A (en) * | 2016-04-26 | 2016-09-07 | 中国科学院地理科学与资源研究所 | Method for estimating instantaneous surface emissivity from passive microwave data |
| CN108445464A (en) * | 2018-03-12 | 2018-08-24 | 南京恩瑞特实业有限公司 | Satellite radar inverting fusion methods of the NRIET based on machine learning |
| CN113486581A (en) * | 2021-07-02 | 2021-10-08 | 宁夏大学 | Machine learning earth surface temperature estimation method, system, storage medium and equipment based on knowledge driving and logical reasoning |
| CN114265836A (en) * | 2021-11-16 | 2022-04-01 | 南京信息工程大学 | All-weather assimilation method and device of satellite microwave hygrothermograph based on cloud area temperature and humidity profile inversion |
| CN114581791A (en) * | 2022-03-02 | 2022-06-03 | 中国农业科学院农业资源与农业区划研究所 | Inversion method and system for atmospheric water vapor content based on MODIS data |
| CN115422703A (en) * | 2022-07-19 | 2022-12-02 | 南京航空航天大学 | Surface thermal infrared emissivity inversion method based on MODIS data and Transformer network |
| US20230131036A1 (en) * | 2021-10-26 | 2023-04-27 | Wuhan University | Retrieval method for surface ozone based on surface ultraviolet radiation irradiance |
| CN116243406A (en) * | 2023-03-23 | 2023-06-09 | 南京信息工程大学 | Full-space assimilation method for satellite infrared radiation rate data |
| CN116467854A (en) * | 2023-03-28 | 2023-07-21 | 复旦大学 | Multi-channel cloud physical characteristic inversion algorithm based on radiation mode and machine learning |
| CN117194957A (en) * | 2023-10-08 | 2023-12-08 | 贵州北盘江电力股份有限公司马马崖光伏分公司 | Ultra-short-term prediction method based on satellite inversion radiation data technology |
| WO2024021225A1 (en) * | 2022-07-29 | 2024-02-01 | 知天(珠海横琴)气象科技有限公司 | High-resolution true-color visible light model generation method, high-resolution true-color visible light model inversion method, and system |
-
2024
- 2024-03-20 CN CN202410319958.0A patent/CN117933095B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101295022A (en) * | 2008-06-25 | 2008-10-29 | 中国农业科学院农业资源与农业区划研究所 | Method for ground surface temperature and emissivity inversion by remote sensing data ASTER |
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
| CN105930664A (en) * | 2016-04-26 | 2016-09-07 | 中国科学院地理科学与资源研究所 | Method for estimating instantaneous surface emissivity from passive microwave data |
| CN108445464A (en) * | 2018-03-12 | 2018-08-24 | 南京恩瑞特实业有限公司 | Satellite radar inverting fusion methods of the NRIET based on machine learning |
| CN113486581A (en) * | 2021-07-02 | 2021-10-08 | 宁夏大学 | Machine learning earth surface temperature estimation method, system, storage medium and equipment based on knowledge driving and logical reasoning |
| US20230131036A1 (en) * | 2021-10-26 | 2023-04-27 | Wuhan University | Retrieval method for surface ozone based on surface ultraviolet radiation irradiance |
| CN114265836A (en) * | 2021-11-16 | 2022-04-01 | 南京信息工程大学 | All-weather assimilation method and device of satellite microwave hygrothermograph based on cloud area temperature and humidity profile inversion |
| CN114581791A (en) * | 2022-03-02 | 2022-06-03 | 中国农业科学院农业资源与农业区划研究所 | Inversion method and system for atmospheric water vapor content based on MODIS data |
| CN115422703A (en) * | 2022-07-19 | 2022-12-02 | 南京航空航天大学 | Surface thermal infrared emissivity inversion method based on MODIS data and Transformer network |
| WO2024021225A1 (en) * | 2022-07-29 | 2024-02-01 | 知天(珠海横琴)气象科技有限公司 | High-resolution true-color visible light model generation method, high-resolution true-color visible light model inversion method, and system |
| CN116243406A (en) * | 2023-03-23 | 2023-06-09 | 南京信息工程大学 | Full-space assimilation method for satellite infrared radiation rate data |
| CN116467854A (en) * | 2023-03-28 | 2023-07-21 | 复旦大学 | Multi-channel cloud physical characteristic inversion algorithm based on radiation mode and machine learning |
| CN117194957A (en) * | 2023-10-08 | 2023-12-08 | 贵州北盘江电力股份有限公司马马崖光伏分公司 | Ultra-short-term prediction method based on satellite inversion radiation data technology |
Non-Patent Citations (4)
| Title |
|---|
| XUSHENG YAN 等: "A 3-D Cloud Detection Method for FY-4A GIIRS and Its Application in Operational Numerical Weather Prediction System", IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 4 September 2023 (2023-09-04) * |
| XUSHENG YAN等: "A 3-D Cloud Detection Method for FY-4A GIIRS and Its Application in Operational Numerical Weather Prediction System", IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 4 September 2023 (2023-09-04), pages 1 - 13 * |
| 漆成莉: "FY-3A气象卫星红外分光计透射率计算及大气参数模拟反演试验", 中国优秀博硕士学位论文全文数据库 (硕士)工程科技Ⅱ辑, 15 December 2004 (2004-12-15) * |
| 金大智 等: "红外高光谱资料近地面通道应用技术研究:海洋地区观测资料同化试验", 气象, no. 06, 21 June 2013 (2013-06-21) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115046894A (en) * | 2022-05-18 | 2022-09-13 | 中国地质大学（武汉） | Cloud haze detection method based on radiation transmission simulation coupling machine learning technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117933095B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109580003B (en) | Method for estimating near-ground atmospheric temperature by thermal infrared data of stationary meteorological satellite | |
| CN117933095B (en) | Earth surface emissivity real-time inversion and assimilation method based on machine learning | |
| Bengtsson et al. | FGGE 4-dimensional data assimilation at ECMWF (weather forecasts). | |
| CN112348292B (en) | Short-term wind power prediction method and system based on deep learning network | |
| CN105243435B (en) | A kind of soil moisture content prediction technique based on deep learning cellular Automation Model | |
| CN114265836A (en) | All-weather assimilation method and device of satellite microwave hygrothermograph based on cloud area temperature and humidity profile inversion | |
| CN110619432A (en) | Feature extraction hydrological forecasting method based on deep learning | |
| CN115204618B (en) | CCMVS region carbon source sink equalization inversion method | |
| CN110909447B (en) | High-precision short-term prediction method for ionization layer region | |
| CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
| CN112329334B (en) | MWHTS and MWTS-II fusion inversion sea surface air pressure method based on simulated bright temperature | |
| CN112100922A (en) | Wind resource prediction method based on WRF and CNN convolutional neural network | |
| Chen et al. | Proactive quality control: Observing system simulation experiments with the Lorenz’96 model | |
| Hu et al. | An improved model for soil surface temperature from air temperature in permafrost regions of Qinghai-Xizang (Tibet) Plateau of China | |
| CN103279030B (en) | Dynamic soft measuring modeling method and device based on Bayesian frame | |
| CN114595876A (en) | Regional wind field prediction model generation method and device and electronic equipment | |
| CN114417728A (en) | Near-surface air temperature inversion method based on temperature, emissivity and deep learning | |
| CN115952743B (en) | Multi-source precipitation data collaborative downscaling method and system coupled with random forest and HASM | |
| CN115510732B (en) | Shelter infrared characteristic simulation rapid algorithm based on deep learning | |
| CN117114176A (en) | Land utilization change prediction method and system based on data analysis and machine learning | |
| CN115840908B (en) | Method for constructing PM2.5 three-dimensional dynamic monitoring field based on microwave link of LSTM model | |
| CN114371519B (en) | Foundation microwave radiometer atmospheric temperature and humidity profile inversion method based on unbiased bright temperature | |
| CN110059972A (en) | Day solar radiation stock assessment method based on functional deepness belief network | |
| CN110019167A (en) | Long-term new forms of energy resource data base construction method and system in one kind | |
| CN115855272A (en) | Method for generating near-real-time hourly all-weather surface temperature |
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 |