CN114417728A - Near-surface air temperature inversion method based on temperature, emissivity and deep learning - Google Patents
Near-surface air temperature inversion method based on temperature, emissivity and deep learning Download PDFInfo
- Publication number
- CN114417728A CN114417728A CN202210098610.4A CN202210098610A CN114417728A CN 114417728 A CN114417728 A CN 114417728A CN 202210098610 A CN202210098610 A CN 202210098610A CN 114417728 A CN114417728 A CN 114417728A
- Authority
- CN
- China
- Prior art keywords
- data
- air temperature
- surface air
- temperature
- inversion
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 230000005540 biological transmission Effects 0.000 claims abstract description 21
- 230000005855 radiation Effects 0.000 claims abstract description 21
- 238000004088 simulation Methods 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 6
- 238000005259 measurement Methods 0.000 claims description 13
- 238000003908 quality control method Methods 0.000 claims description 12
- 238000011160 research Methods 0.000 claims description 11
- 238000012950 reanalysis Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 abstract description 3
- 230000009471 action Effects 0.000 abstract description 2
- 230000007246 mechanism Effects 0.000 abstract description 2
- 230000006872 improvement Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000000443 aerosol Substances 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000003915 air pollution Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004071 biological effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000029553 photosynthesis Effects 0.000 description 1
- 238000010672 photosynthesis Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000005068 transpiration Effects 0.000 description 1
Images
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
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- 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
-
- 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
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Mathematical Analysis (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Molecular Biology (AREA)
- Computational Mathematics (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention provides a near-surface air temperature inversion method based on temperature, emissivity and deep learning, which comprises four steps of establishing a thermal radiation transmission equation, establishing an expert knowledge base, establishing a high-precision database, calculating an inversion result and verifying, wherein the expert knowledge base is established based on the thermal radiation transmission equation and priori knowledge, parameters required for inversion are determined through a radiation transmission mechanism, from the physical perspective, near-surface heat interaction influence is considered, the surface temperature and emissivity are used as the priori knowledge, and the simulation data and the acquired high-precision data are used for realizing higher inversion precision, so that the advantages of a physical model and deep learning are fully utilized and combined, a high-precision near-surface air temperature inversion result can be obtained, a DL-NN algorithm is adopted for processing the ill-state problem of near-surface air temperature inversion and processing the nonlinear relation between the near-surface air temperature and the atmospheric average action temperature in different seasons and regions, the inversion precision is improved, and the universality is improved.
Description
Technical Field
The invention relates to the technical field of near-surface air temperature inversion, in particular to a near-surface air temperature inversion method based on temperature, emissivity and deep learning.
Background
The near-surface air temperature generally refers to the atmospheric temperature about 2m away from the ground, is an important parameter for researching the interaction between an area and a global atmosphere-earth system, and relates to the field of research requiring multiple correlations, such as global climate change, hydrology, aerology, ecology, agricultural production, urban heat island effect, air pollution research and the like, wherein the near-surface air temperature is used as an input parameter, high-precision inversion of the near-surface air temperature can better understand the climate change or local interference and simulate a complex ground process, and the near-surface air temperature also plays an important role in physical processes such as evaporation transpiration, photosynthesis, heat transfer and the like, so that the space-time pattern for accurately estimating the near-surface air temperature has important significance for further understanding the ecological, hydrology, climate, agriculture and terrestrial biological activities;
the meteorological station is one of the most common ways to acquire the observation data of the near-surface air temperature, and has the advantages that the data of the sample points with different time resolution and high precision of a long time sequence can be acquired, however, the near-surface air temperature is influenced by the exchange process among three earth system units of land/sea and atmosphere, the spatial pattern distribution of the near-surface air temperature is different and very complex, the surface relief of a plurality of regions is large, the surface types are various, the meteorological stations are not distributed uniformly, particularly, the observation stations are lacked in remote and complex regions, therefore, limited spatial pattern information can be provided only on a regional scale, in addition, near-surface temperature data which is continuous in time and space is usually required in the climate change research, and the recording mode of sample point data restricts the precision of simulation analysis of related climate models and limits the development of geoscience models to a certain extent, thereby influencing the understanding of people on global climate and environmental change;
the current methods for obtaining the near-surface air temperature of space-time continuous distribution mainly comprise three methods: the first method is a spatial interpolation method based on a meteorological observation station; the second method is the atmospheric vertical profile method; the third method is based on a remote sensing method, however, in the methods, ground measurement data are limited by the influence of insufficient space representativeness and ground heterogeneity, and continuous space-time data with large scale cannot be provided.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a near-surface air temperature inversion method based on temperature, emissivity and deep learning, so as to solve the problem of low universality in the prior art.
In order to realize the purpose of the invention, the invention is realized by the following technical scheme: the near-surface air temperature inversion method based on temperature, emissivity and deep learning comprises the following steps:
step one, establishing a thermal radiation transmission equation
Determining a research area, acquiring satellite remote sensing sensor data and ERA5-Land data of the research area, establishing a model data set according to the acquired satellite remote sensing sensor data and ERA5-Land data, introducing auxiliary data into the model data set, and establishing a thermal radiation transmission equation from the perspective of the satellite remote sensing sensor data;
step two, constructing an expert knowledge base
Constructing an expert knowledge base based on the thermal radiation transmission equation established in the step one and combining the prior knowledge, and then carrying out physical logic reasoning through the constructed expert knowledge base to deduce parameters required by the inversion of the near-surface air temperature, namely analyzing factors influencing the near-surface air temperature in the energy radiation transmission process through the physical logic reasoning;
step three, constructing a high-precision database
Performing double-layer quality control by using sensor data in the model data set in the step one in combination with ERA5-Land data, and constructing a database for high-precision inversion of near-surface air temperature;
step four: computing inversion results and validating
And (4) performing optimization calculation by using a deep learning algorithm DL-NN according to the high-precision inversion near-surface air temperature database constructed in the third step to obtain a model inversion result, and performing comparison verification on the model inversion result and the ERA5-Land reanalysis data, the meteorological site data and the simulation data in the model data set in the first step.
The further improvement lies in that: in the first step, the auxiliary data comprises simulation data, measurement data and high-precision assimilation data, the simulation data is simulation data of an atmosphere simulation program, the measurement data is field measurement data from a meteorological satellite observation station, and the high-precision assimilation data is fully verified product data.
The further improvement lies in that: in the first step, due to the time-space deviation of data acquisition, abnormal values and unrepresentative data need to be discharged, and the data in the database can effectively drive the model to reflect the relation between the parameters and the near-surface air temperature.
The further improvement lies in that: in the second step, the difference between the surface temperature and the near-surface air temperature is influenced by the surface energy exchange, and then the surface temperature and the emissivity are added as priori knowledge.
The further improvement lies in that: in the second step, the factors influencing the near-surface air temperature comprise the satellite brightness temperature, the water vapor and the surface emissivity.
The further improvement lies in that: the double-layer quality control comprises a first layer control and a second layer control, wherein the first layer control utilizes a sensor ground surface temperature product to carry out clearance control, and the second layer control utilizes the sensor ground surface temperature in combination with ERA5-Land to analyze ground surface temperature data to carry out second layer clearance control.
The further improvement lies in that: the database construction mode comprises the following steps:
s1: based on MYD11A1 data pixel quality filtering, identifying a low-quality pixel value, regarding a missing pixel and the low-quality pixel as invalid pixels, searching a corresponding pixel in MYD021KM data according to the position information of the invalid pixels, and removing the invalid pixels;
s2: and performing time linear interpolation and spatial geographic registration on the ERA5-land reanalysis data, and finishing the database construction work when the difference between the surface temperature of the two is less than a preset value, namely the collected data after quality control is considered to be correct.
The further improvement lies in that: and in the fourth step, the data in the database is adjusted according to the accuracy of the comparison verification result until the requirement of the accuracy is met.
The invention has the beneficial effects that: according to the near-surface air temperature inversion method based on temperature, emissivity and deep learning, an expert knowledge base is established based on a thermal radiation transmission equation and priori knowledge, parameters required for inversion are determined through a radiation transmission mechanism, from the physical perspective, near-surface heat interaction influence is considered, the surface temperature and emissivity are used as the priori knowledge, and simulation data and collected high-precision data are used for achieving high inversion precision, so that the advantages of a physical model and the deep learning are fully utilized and combined, a high-precision near-surface air temperature inversion result can be obtained, and the universality of the method can be improved.
Drawings
FIG. 1 is a flowchart illustrating a first step of the present invention.
Fig. 2 is a schematic diagram of a near-surface air temperature inversion process according to a first embodiment of the present invention.
Detailed Description
In order to further understand the present invention, the following detailed description will be made with reference to the following examples, which are only used for explaining the present invention and are not to be construed as limiting the scope of the present invention.
Example one
According to fig. 1, the embodiment provides a near-surface air temperature inversion method based on temperature, emissivity and depth learning, which includes the following steps:
step one, establishing a thermal radiation transmission equation
Determining a research area, acquiring satellite remote sensing sensor data and ERA5-Land data of the research area, establishing a model data set according to the acquired satellite remote sensing sensor data and ERA5-Land data, introducing auxiliary data into the model data set, and establishing a thermal radiation transmission equation from the perspective of the satellite remote sensing sensor data, wherein the ERA5-Land data is generated by reconstructing ERA5 climate in the terrestrial part of an analysis data set by the medium-term weather forecast center in Europe. Provide decades of terrestrial variable reanalysis data at high resolution;
step two, constructing an expert knowledge base
Constructing an expert knowledge base based on the thermal radiation transmission equation established in the step one and combining the prior knowledge, and then carrying out physical logic reasoning through the constructed expert knowledge base to deduce parameters required by the inversion of the near-surface air temperature, namely analyzing factors influencing the near-surface air temperature in the energy radiation transmission process through the physical logic reasoning;
step three, constructing a high-precision database
Performing double-layer quality control by using sensor data in the model data set in the step one in combination with ERA5-Land data, and constructing a database for high-precision inversion of near-surface air temperature;
step four: computing inversion results and validating
And performing optimization calculation by using a deep learning algorithm DL-NN according to a high-precision inversion near-surface air temperature database constructed in the third step to obtain a model inversion result, and performing comparison verification on the model inversion result and the ERA5-Land reanalysis data, meteorological station data and simulation data in the model data set in the first step, wherein the DL-NN algorithm is adopted to process the ill-condition problem of near-surface air temperature inversion and process the complex nonlinear relation between the near-surface air temperature and the atmospheric average action temperature in different seasons and regions, so that the inversion precision is improved, and the universality is improved.
In the first step, the auxiliary data comprises simulation data, measurement data and high-precision assimilation data, the simulation data is simulation data of an atmospheric simulation program, the measurement data is field measurement data from a meteorological satellite observation station, and the high-precision assimilation data is fully verified product data.
In the first step, due to the time-space deviation of data acquisition, abnormal values and unrepresentative data need to be discharged, and the data in the database can effectively drive the model to reflect the relation between the parameters and the near-surface air temperature.
In the second step, because the surface energy exchange influences the difference between the surface temperature and the near-surface air temperature, the surface temperature and the emissivity are added as priori knowledge.
And in the second step, factors influencing the near-surface air temperature comprise the satellite brightness temperature, water vapor and the surface emissivity.
The double-layer quality control comprises a first layer control and a second layer control, wherein the first layer control utilizes a sensor ground surface temperature product to carry out clearance checking, the sensor ground surface temperature data are obtained from a thermal infrared wave band under a clear air condition and comprise a plurality of deficiency values and low quality values caused by cloud, aerosol and other factors, and the second layer control utilizes the sensor ground surface temperature to be combined with ERA5-Land to analyze the ground surface temperature data again to carry out the second layer clearance checking.
The database construction mode comprises the following steps:
s1: based on MYD11A1 data pixel quality filtering, identifying a low-quality pixel value, regarding a missing pixel and the low-quality pixel as invalid pixels, searching a corresponding pixel in MYD021KM data according to the position information of the invalid pixels, and removing the invalid pixels;
s2: and performing time linear interpolation and spatial geographic registration on the ERA5-land reanalysis data, and finishing the database construction work when the difference between the surface temperature of the two is less than a preset value, namely the collected data after quality control is considered to be correct.
In this embodiment, in order to construct and evaluate the proposed model accuracy and improve the applicability of the model, the data in the database is divided into three groups according to different situations, as shown in table one:
watch 1
Wherein, No. 1 is BTs and LSE, LST (surface temperature) with water vapor and thermal infrared bands 29, 31 and 32 suitable for inversion in daytime; number 2 is BTs and LSE, LST (surface temperature) of thermal infrared bands 29, 31, 32 and 33 and infrared bands 20, 22 and 23 suitable for evening inversion; no. 3 is BTs and LSE, LST (surface temperature) of thermal infrared bands 29, 31, 32, and 33 suitable for inversion in the daytime and at night, model accuracy is evaluated according to inversion results of three groups of different conditions, and as shown in fig. 2, the flowchart of the whole inversion model is shown, and parameter setting is performed on training data and test data by using simulation data of an atmospheric simulation program (MODTRAN simulates MODIS data), such as: LST (surface temperature) (270K-330K), WVC (0.1-4.5g/cm2), LSE (vegetation, water, soil and rock).
And the accuracy of the result needs to be verified and analyzed, and according to the above inversion case No. 1, as shown in table two:
watch two
The second table summarizes inversion errors of LST (surface temperature) and LSE (surface emissivity) in the inversion case No. 1, and it is found from the data in the second table that when the number of hidden layers is 9 and the hidden node is 600, the inversion accuracy of LST (surface temperature) is the highest (MAE is 0.59 and RMSE is 0.93), MAE is the average absolute error, and RMSE is the root mean square error.
In the case of the above inversion No. 2, as shown in table three:
watch III
Table three summarizes the inversion errors of LST and LSE in the case of inversion No. 2, and when the number of hidden layers is 7 and the hidden node is 500, the inversion LST has the highest accuracy (MAE ═ 0.43, RMSE ═ 0.65).
The inversion case 3 is shown in table four:
watch four
Wherein, table four summarizes the inversion errors of LST and LSE in the reaction case No. 3, and when the number of hidden layers is 7 and the hidden node is 500, the inversion LST has the highest accuracy (MAE ═ 0.43, RMSE ═ 0.65).
According to data, the accuracy of inversion of the LST (surface temperature) is obviously improved by adding water vapor information to the surface temperature compared with combination of day and night.
Example two
The embodiment provides a near-surface air temperature inversion method based on temperature, emissivity and deep learning, which comprises the following steps:
step one, establishing a thermal radiation transmission equation
Determining a research area, acquiring satellite remote sensing sensor data and ERA5-Land data of the research area, establishing a model data set according to the acquired satellite remote sensing sensor data and ERA5-Land data, introducing auxiliary data into the model data set, and establishing a thermal radiation transmission equation from the perspective of the satellite remote sensing sensor data;
step two, constructing an expert knowledge base
Constructing an expert knowledge base based on the thermal radiation transmission equation established in the step one and combining the prior knowledge, and then carrying out physical logic reasoning through the constructed expert knowledge base to deduce parameters required by the inversion of the near-surface air temperature, namely analyzing factors influencing the near-surface air temperature in the energy radiation transmission process through the physical logic reasoning;
step three, constructing a high-precision database
Performing double-layer quality control by using sensor data in the model data set in the step one in combination with ERA5-Land data, and constructing a database for high-precision inversion of near-surface air temperature;
step four: computing inversion results and validating
And (4) performing optimization calculation by using a deep learning algorithm DL-NN according to the high-precision inversion near-surface air temperature database constructed in the third step to obtain a model inversion result, and performing comparison verification on the model inversion result and the ERA5-Land reanalysis data, the meteorological site data and the simulation data in the model data set in the first step.
In the first step, the auxiliary data comprises simulation data, measurement data and high-precision assimilation data, the simulation data is simulation data of an atmospheric simulation program, the measurement data is field measurement data from a meteorological satellite observation station, and the high-precision assimilation data is fully verified product data.
In the first step, due to the time-space deviation of data acquisition, abnormal values and unrepresentative data need to be discharged, and the data in the database can effectively drive the model to reflect the relation between the parameters and the near-surface air temperature.
In the second step, because the surface energy exchange influences the difference between the surface temperature and the near-surface air temperature, the surface temperature and the emissivity are added as priori knowledge.
And in the second step, factors influencing the near-surface air temperature comprise the satellite brightness temperature, water vapor and the surface emissivity.
The double-layer quality control comprises a first layer control and a second layer control, wherein the first layer control utilizes the sensor ground surface temperature product to perform clearance control, and the second layer control utilizes the sensor ground surface temperature combined with ERA5-Land to analyze ground surface temperature data to perform second layer clearance control.
The database construction mode comprises the following steps:
s1: based on MYD11A1 data pixel quality filtering, identifying a low-quality pixel value, regarding a missing pixel and the low-quality pixel as invalid pixels, searching a corresponding pixel in MYD021KM data according to the position information of the invalid pixels, and removing the invalid pixels;
s2: and performing time linear interpolation and spatial geographic registration on the ERA5-land reanalysis data, and finishing the database construction work when the difference between the surface temperature of the two is less than a preset value, namely the collected data after quality control is considered to be correct.
And in the fourth step, adjusting the data in the database according to the accuracy of the comparison verification result until the requirement of the accuracy is met.
The difference between the first embodiment and the second embodiment is that when the accuracy of the comparison verification result is not satisfactory, the established high-accuracy database is supplemented with the reliable data, which may be ground station data, mature official product data (for example, MODIS official product and ERA5-Land reanalysis data) or other reliable data (for example, MODTRAN simulation data), and then the training and testing database is updated until the inverted and updated data are iterated repeatedly until the accuracy requirement is met, which is a cyclic iteration process, wherein the brightness temperature information in the high-accuracy database is used for simultaneously inverting the surface temperature and the surface emissivity, and then the inverted surface temperature and the surface emissivity are used as the prior knowledge for inverting the near-surface temperature. And after the two inversion accuracies meet the requirements, one near-surface air temperature inversion is regarded as being completed.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (8)
1. The near-surface air temperature inversion method based on temperature, emissivity and deep learning is characterized by comprising the following steps of: the method comprises the following steps:
step one, establishing a thermal radiation transmission equation
Determining a research area, acquiring satellite remote sensing sensor data and ERA5-Land data of the research area, establishing a model data set according to the acquired satellite remote sensing sensor data and ERA5-Land data, introducing auxiliary data into the model data set, and establishing a thermal radiation transmission equation from the perspective of the satellite remote sensing sensor data;
step two, constructing an expert knowledge base
Constructing an expert knowledge base based on the thermal radiation transmission equation established in the step one and combining the prior knowledge, and then carrying out physical logic reasoning through the constructed expert knowledge base to deduce parameters required by the inversion of the near-surface air temperature, namely analyzing factors influencing the near-surface air temperature in the energy radiation transmission process through the physical logic reasoning;
step three, constructing a high-precision database
Performing double-layer quality control by using sensor data in the model data set in the step one in combination with ERA5-Land data, and constructing a database for high-precision inversion of near-surface air temperature;
step four: computing inversion results and validating
And (4) performing optimization calculation by using a deep learning algorithm DL-NN according to the high-precision inversion near-surface air temperature database constructed in the third step to obtain a model inversion result, and performing comparison verification on the model inversion result and the ERA5-Land reanalysis data, the meteorological site data and the simulation data in the model data set in the first step.
2. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: in the first step, the auxiliary data comprises simulation data, measurement data and high-precision assimilation data, the simulation data is simulation data of an atmosphere simulation program, the measurement data is field measurement data from a meteorological satellite observation station, and the high-precision assimilation data is fully verified product data.
3. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: in the first step, due to the time-space deviation of data acquisition, abnormal values and unrepresentative data need to be discharged, and the data in the database can effectively drive the model to reflect the relation between the parameters and the near-surface air temperature.
4. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: in the second step, the difference between the surface temperature and the near-surface air temperature is influenced by the surface energy exchange, and then the surface temperature and the emissivity are added as priori knowledge.
5. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: in the second step, the factors influencing the near-surface air temperature comprise the satellite brightness temperature, the water vapor and the surface emissivity.
6. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: the double-layer quality control comprises a first layer control and a second layer control, wherein the first layer control utilizes a sensor ground surface temperature product to carry out clearance control, and the second layer control utilizes the sensor ground surface temperature in combination with ERA5-Land to analyze ground surface temperature data to carry out second layer clearance control.
7. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: the database construction mode comprises the following steps:
s1: based on MYD11A1 data pixel quality filtering, identifying a low-quality pixel value, regarding a missing pixel and the low-quality pixel as invalid pixels, searching a corresponding pixel in MYD021KM data according to the position information of the invalid pixels, and removing the invalid pixels;
s2: and performing time linear interpolation and spatial geographic registration on the ERA5-land reanalysis data, and finishing the database construction work when the difference between the surface temperature of the two is less than a preset value, namely the collected data after quality control is considered to be correct.
8. The near-surface air temperature inversion method based on temperature and emissivity and deep learning of claim 1, wherein the near-surface air temperature inversion method comprises the following steps: and in the fourth step, the data in the database is adjusted according to the accuracy of the comparison verification result until the requirement of the accuracy is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210098610.4A CN114417728A (en) | 2022-01-27 | 2022-01-27 | Near-surface air temperature inversion method based on temperature, emissivity and deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210098610.4A CN114417728A (en) | 2022-01-27 | 2022-01-27 | Near-surface air temperature inversion method based on temperature, emissivity and deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114417728A true CN114417728A (en) | 2022-04-29 |
Family
ID=81280181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210098610.4A Pending CN114417728A (en) | 2022-01-27 | 2022-01-27 | Near-surface air temperature inversion method based on temperature, emissivity and deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114417728A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN114838827A (en) * | 2022-05-23 | 2022-08-02 | 河北地质大学 | Earth surface temperature inversion channel selection method based on MERSI-II remote sensing data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652404A (en) * | 2020-02-06 | 2020-09-11 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| 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 |
-
2022
- 2022-01-27 CN CN202210098610.4A patent/CN114417728A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652404A (en) * | 2020-02-06 | 2020-09-11 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| HAN WANG等: ""A method for land surface temperature retrieval based onmodel-data-knowledge-driven and deep learning", 《REMOTE SENSING OF ENVIRONMENT》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN113486581B (en) * | 2021-07-02 | 2024-08-16 | 宁夏大学 | Knowledge-driven and logic-reasoning-based machine learning earth surface temperature estimation method, system, storage medium and equipment |
| CN114838827A (en) * | 2022-05-23 | 2022-08-02 | 河北地质大学 | Earth surface temperature inversion channel selection method based on MERSI-II remote sensing data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109580003B (en) | Method for estimating near-ground atmospheric temperature by thermal infrared data of stationary meteorological satellite | |
| US11333796B2 (en) | Spatial autocorrelation machine learning-based downscaling method and system of satellite precipitation data | |
| CN112684520B (en) | Weather forecast correction method and device, computer equipment and storage medium | |
| CN110751094B (en) | Crop yield estimation method based on GEE comprehensive remote sensing image and deep learning method | |
| CN109142679B (en) | Forest soil nutrient space prediction method based on artificial neural network kriging interpolation | |
| CN112163375B (en) | Long-time sequence near-ground ozone inversion method based on neural network | |
| CN114417728A (en) | Near-surface air temperature inversion method based on temperature, emissivity and deep learning | |
| CN112699959B (en) | Multi-source multi-scale precipitation data fusion method and device based on energy functional model | |
| CN109784552A (en) | A kind of construction method of the space variable coefficient PM2.5 concentration appraising model based on Re-ESF algorithm | |
| CN106483147B (en) | Long-time sequence passive microwave soil moisture precision improvement research method based on multi-source data | |
| CN114005048A (en) | Multi-temporal data-based land cover change and thermal environment influence research method | |
| CN115630567A (en) | Coastal zone soil organic carbon reserve simulation and prediction method | |
| CN113408111B (en) | Atmospheric precipitation inversion method and system, electronic equipment and storage medium | |
| CN116297068A (en) | Aerosol optical thickness inversion method and system based on earth surface reflectivity optimization | |
| CN114595876A (en) | Regional wind field prediction model generation method and device and electronic equipment | |
| CN105046046A (en) | Ensemble Kalman filter localization method | |
| CN117933095B (en) | Earth surface emissivity real-time inversion and assimilation method based on machine learning | |
| CN115859789A (en) | Method for improving inversion accuracy of polar atmosphere temperature profile | |
| Park et al. | Improvement of satellite-based estimation of gross primary production through optimization of meteorological parameters and high resolution land cover information at regional scale over East Asia | |
| CN114490905A (en) | Clear sky surface net long wave radiation integrated inversion method and system | |
| CN109948175B (en) | Satellite remote sensing albedo missing value inversion method based on meteorological data | |
| CN115937714A (en) | Net Primary Productivity Estimation Method Based on Neural Network and CASA Model | |
| CN114169215A (en) | Surface temperature inversion method coupling remote sensing and regional meteorological model | |
| CN109871552B (en) | Soil temperature estimation method for different depths | |
| Xia et al. | A Markov-Kalman model of land-use change prediction in XiuHe Basin, China |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220429 |
|
| RJ01 | Rejection of invention patent application after publication |