CN116341352B - Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint - Google Patents
Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint Download PDFInfo
- Publication number
- CN116341352B CN116341352B CN202210892133.9A CN202210892133A CN116341352B CN 116341352 B CN116341352 B CN 116341352B CN 202210892133 A CN202210892133 A CN 202210892133A CN 116341352 B CN116341352 B CN 116341352B
- Authority
- CN
- China
- Prior art keywords
- surface temperature
- data
- temperature
- observation
- station
- 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.)
- Active
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000003068 static effect Effects 0.000 title claims abstract description 25
- 230000005855 radiation Effects 0.000 claims abstract description 20
- 230000005540 biological transmission Effects 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000003908 quality control method Methods 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims description 26
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- 238000011478 gradient descent method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000012876 topography Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 6
- 102100033668 Cartilage matrix protein Human genes 0.000 description 4
- 101001018382 Homo sapiens Cartilage matrix protein Proteins 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000191 radiation effect Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 238000012067 mathematical method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- 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
Abstract
The invention discloses a static satellite land infrared bright temperature simulation method constrained by earth surface temperature observation information, which comprises the following steps: s1, acquiring data of numerical weather mode forecast and ground surface temperature data of live observation of a reference station, and preprocessing; s2, performing quality control on the data; s3, assimilating the two types of data; s4, extracting the atmospheric elements and the earth surface elements corresponding to each horizontal space grid point, using an earth surface emissivity data set changing month by month as the input of a radiation transmission model, simulating the bright temperature of an infrared channel of the static satellite infrared imager, and then converting the simulated bright temperature value corresponding to the horizontal space grid point into the pixel point of the static satellite infrared imager to complete the static satellite terrestrial infrared bright temperature simulation. The invention can improve the infrared bright temperature simulation level of the static satellite infrared imager data, especially the simulation level in the land area.
Description
Technical Field
The invention relates to the field of atmospheric science research, in particular to a static satellite land infrared bright temperature simulation method considering earth surface temperature observation information constraint.
Background
With the continuous development of meteorological satellite observation technology and numerical weather forecast, satellite data has been widely applied to global and regional numerical weather forecast modes, wherein stationary satellites play an important role. Compared with the traditional satellite infrared detector, the new generation of static satellite infrared imaging device represented by FY-4A/AGRI, GOES-R/ABI in the United states and Himaware-8/AHI in Japan has the advantages that the horizontal detection resolution can reach 2km at most, the time resolution can reach 10 minutes, and the observation channel is sensitive to the atmosphere and the surface information of different layers. The high space-time resolution and channel advantage make it an important observation information source in numerical weather forecast. However, in order to effectively assimilate these channel observations in numerical mode, it is first necessary to simulate the channel bright temperature of a satellite infrared imager based on atmospheric elements, a process called infrared bright temperature simulation.
The satellite bright temperature simulation under the clear air atmosphere condition adopts the mode that the atmospheric temperature, water vapor, air pressure, earth surface information, geometric parameters corresponding to a satellite view field and the like are input into a radiation transmission model, the atmospheric radiation effect, the earth surface radiation effect, the solar short wave radiation effect and the like are calculated, the radiation entering the satellite sensor upwards through the top of an atmospheric layer is simulated by a radiation transmission equation, and the radiation is converted into bright temperature. Taking himaware-8 satellite AHI imager as an example, the weighting functions of channels of 8.60 μm,10.45 μm,11.20 μm,12.35 μm and 13.30 μm are near the ground and sensitive to the ground surface state, and the performance of bright temperature simulation on the ground is greatly dependent on the description of the ground surface emissivity and the ground surface temperature of the radiation transmission model, which is dependent on the characterization capability of the radiation transmission model for the infrared band ground surface emissivity of different underlying surface types on one hand, and on the other hand, is also strongly dependent on the ground surface temperature information of the input radiation transmission model.
Reviewing the existing infrared imager bright temperature simulation scheme, students at home and abroad often adopt quick radiation modes such as CRTM (Weng et al 2007), RTTOV (Hocking et al 2020), ARMS (Wenget et al 2021) and the like as models, and atmospheric and surface elements predicted by numerical modes are used as model input quantities to establish an infrared radiation simulation scheme. Zou and Zhuge (2016) use ECMWF mode data as input fields, and use CRTM and RTTOV to simulate Himaware-8/AHI infrared imager channel brightness temperature under clear sky conditions; tang et al (2021) uses GFS mode data as an input field to simulate the channel brightness temperature of an FY-4A/AGRI infrared imager on the ocean surface by ARMS; zhang et al (2016) uses CRTM to simulate the channel brightness temperature of GOES-R/ABI infrared imagers from WRF mesoscale numerical mode data as an input field.
However, these studies have limitations, and are remarkable in that, as the surface temperature distribution characteristics of the numerical weather model forecast have great uncertainty compared with the observation (Zheng et al 2012), the radiation transmission model can have great errors in the calculated surface radiation, so that the near-ground channel infrared radiation simulation effect influenced by the surface is not ideal, and the simulation accuracy of the land surface is obviously lower than that of the ocean surface. In view of this, satellite infrared bright temperature simulation schemes on land need to be further improved compared to satellite infrared bright temperature simulation of the ocean area.
One way to enhance the ground satellite ground surface sensitive channel bright temperature simulation is to improve the ground surface temperature information of the input radiation transmission model, and one theoretically more effective way is to modify the ground surface temperature structure of the numerical mode forecast by combining live observation information in the existing scheme. For example, 2400 reference stations are distributed densely in the middle eastern region, the real-time observation of the surface temperature of the underlying surface is realized besides the conventional 2-meter-height ground air temperature observation, the acquisition time frequency reaches 5 minutes, and the direct measurement mode has absolute advantages in accuracy compared with the numerical mode prediction. Therefore, the live ground surface temperature observation data of the station can provide powerful support for the infrared bright temperature simulation of the static satellite AHI imager.
Therefore, considering the accuracy problem of the current static satellite infrared bright temperature simulation scheme under land conditions, we consider that the accuracy of the data of the common numerical weather mode is improved by using the more accurate station surface temperature observation information, and an assimilation algorithm of the corresponding mode forecast surface temperature and the station live surface temperature is developed, so that the improved satellite infrared bright temperature simulation scheme is established by using the surface temperature observation information to restrict the mode background field.
Disclosure of Invention
The invention aims to: the invention aims to provide a static satellite land infrared bright temperature simulation method constrained by earth surface temperature observation information, aiming at the defects of the prior art, improving the infrared bright temperature simulation effect of a static satellite infrared imager in a land area and improving the application capability of the static satellite infrared imager in a numerical forecasting mode.
The technical scheme is as follows: the invention relates to a static satellite land infrared bright temperature simulation method constrained by earth surface temperature observation information, which comprises the following steps:
s1, acquiring data of numerical weather mode forecast and ground surface temperature data of live observation of a reference station, and respectively preprocessing the two types of data;
s2, quality control is carried out on the numerical mode data and the surface temperature data which are preprocessed in the step S1;
s3, assimilating the two types of data processed in the step S2, wherein the steps are as follows:
s31, adding a penalty term for station surface temperature observation in the variation assimilation cost function, so as to establish the variation assimilation cost function considering station surface temperature observation, wherein the variation assimilation cost function comprises a background field penalty term, an observation penalty term except the surface temperature and a newly added surface temperature observation penalty term;
s32, adding a control variable of the surface temperature in a background error covariance matrix used in a background field penalty term, and establishing a co-correlation between the surface temperature and the air temperature;
s33, adopting a classical conjugate gradient descent method to solve gradient descent of a newly constructed variation assimilation cost function, converging a conjugate gradient descent iteration algorithm when a numerical value is iterated to 1/1000 of an initial value, and finally, obtaining a value corresponding to the convergence as an optimal analysis field after functional minimization, wherein the optimal analysis field comprises analysis values of an atmospheric element and a surface element;
s4, extracting the atmospheric elements and the earth surface elements corresponding to each horizontal space grid point obtained in the step S3, adopting an earth surface emissivity data set changing month by month as the input of a radiation transmission model, simulating the bright temperature of an infrared channel of the static satellite infrared imager, and then converting the simulated bright temperature value corresponding to the horizontal space grid point into the pixel point of the static satellite infrared imager to complete the static satellite terrestrial infrared bright temperature simulation.
The further preferable technical scheme of the invention is that the preprocessing of the data of the log weather pattern forecast in the step S1 comprises: extracting longitude, latitude, station altitude and surface temperature;
preprocessing of ground surface temperature data live observed by a reference station includes extracting the temperature of the atmosphere, water vapor and air pressure, and ground surface temperature and wind field.
Preferably, the specific content of performing quality control on the two types of data in step S2 includes: significant error checking, background field outlier checking, spatial consistency checking, temporal consistency checking, and station and pattern terrain height consistency checking.
Preferably, the step of significant error checking is:
setting a critical error checking threshold gamma min ＝-50，γ max ＝70；
If it isThen reject +.>Data.
Preferably, the step of background field outlier verification is:
first, the surface temperature of the numerical mode simulation is simulatedPerforming spatial linear interpolation, wherein the superscript b represents a mode, i and j represent horizontal coordinate numbers of the mode, interpolating to spatial positions of ground weather station data, and obtaining the mode surface temperature +.>
Second, estimateObservation errors of ground surface temperature data are counted according to long-time samples, a triangle cap method is adopted, and ground surface temperature data inverted by MODIS satellites are utilized +.>And surface temperature data of GFS analytical field +.>Interpolation to the observation station point position, and calculation of the observation error sigma of the station surface temperature STA ，
wherein ,
then performing outlier verification:
if it isThen reject +.>Data.
Preferably, the step of spatial consistency checking is:
first, the radius of the space consistency test is calculated for each site with a horizontal distance of 220kmMean +.about.within 220km radius from its surroundings>The difference between them, noted as:
then a spatial consistency check is performed:
if it isThen reject +.>Data.
Preferably, the specific steps of the time consistency test are as follows:
first, each site is calculated separately using one hour as a time window for time consistency checkAverage value +.about.1 hour before and after its time>The difference between them, noted as:
then, a time consistency check is performed:
if it isThen reject +.>Data.
Preferably, the specific steps of the height consistency test of the measuring station and the mode terrain are as follows:
first, the topography height H of the numerical mode is interpolated in a bilinear manner b (i, j) interpolating to the spatial position of the ground weather station data to obtain the model terrain height H at the nth ground weather station position b (n) and calculating the station height H o (n) the difference from the pattern terrain height |H o (n)-H m (n)|；
A high consistency check is then performed:
if |H o (n)-H m (n) | > 50m, then cull the nth siteData.
Preferably, the method for assimilating the two types of data in step S3 is as follows:
s31, increasing the observation of the surface temperature of the station in the variation assimilation cost function of the formula (7)To establish a variation assimilation cost function taking into account the observation of the station surface temperature, as in equation (8),
J(x)＝J B +J O (7)；
wherein, a background field penalty term is includedObservation penalty term other than surface temperature +.>And a newly added surface temperature observation penalty termEach penalty term x represents a value comprising the surface temperature analysis value +.>Is the assimilation analytical field, x b Representing the inclusion of surface temperature forecast values->The model background field of (a), H represents an observation operator, O represents an observation error matrix, and B represents a background error covariance matrix;
s32, improving the background error covariance matrix B to B new The background field penalty term becomes
B new To increase the surface temperature T in the original background error covariance B s And establish the surface temperature T s And air temperature T a Is related to B new As in (9),
wherein, psi, χ, T, RH, P s A flow function, a velocity potential function, an air temperature, a relative humidity, and an air pressure, respectively representing an atmospheric state quantity; χ_u, T a_u and Ps U represents the velocity potential function, temperature and air pressure of the unbalanced portion, respectively; t (T) s U is the unbalanced portion of the surface temperature; c (C) ψ,χ RepresentsCorrelation between the velocity potential function χ and the flow function ψ;representing the temperature T a Correlation with the stream function ψ; />Representing the air pressure P s Correlation with the stream function ψ; i represents an identity matrix;
representing the surface temperature T s And air temperature T a Correlation between _u ++>In the form of a function of (c) is,
δT s (i,j)＝δT s _u(i,j)+δT s _b(i,j) (10)；
where i, j, k denote horizontal and vertical coordinates, b denote partitions of different features, δT s B is the equilibrium part of the surface temperature, and is represented by the air temperature T of formula (11) a Linear regression of u represents that the portion embodies the surface temperature T s And air temperature T a Correlation of _u, and obtaining linear correlation coefficient of the two by sample statisticsAccording to the characteristics of regional differences of the surface temperature and the like, different subareas b are established at intervals of 5 degrees of latitude, and features under the condition of different latitudes are described;
realization of pairs by Cholesky decompositionInverting the matrix;
s33, adopting a classical conjugate gradient descent method to construct a gradient of the cost function newly constructed by the formula (8)Gradient decrease, when->Iterating the numerical values to the initial +.>1/1000 of the total number of the steps, converging the conjugate gradient descent iterative algorithm, and finally, the corresponding x is the optimal analysis field after functional minimization and comprises atmospheric elements and earth surface elements>Is a result of the analysis of (a).
The beneficial effects are that: (1) According to the method, an improved land infrared bright temperature simulation scheme is established for an infrared imager of a stationary satellite, the real-time observation data of the ground surface temperature of a meteorological station and Weather Research and Forecasting (WRF) numerical mode data are combined, the constraint of the real-time observation information of the station on a numerical mode background field is achieved through establishing a corresponding data assimilation process flow, the optimized ground surface temperature field is used as an input field of a radiation transmission model, the infrared bright temperature simulation scheme is established, simulation rationality and accuracy are improved, and the infrared bright temperature simulation level of the data of the infrared imager of the stationary satellite, particularly the simulation level in a land area, can be improved through the method.
(2) The invention adopts variation analysis as a mathematical method of an assimilation algorithm, and is based on the traditional three-dimensional variation functional, and compared with the traditional method, the improvement is that a cost function penalty term related to station ground surface temperature observation data is constructed, and an error covariance model of the atmosphere temperature and the ground surface temperature is constructed in a background error covariance. Solving an assimilation analysis field of the surface temperature through variation minimization iteration, so that correction of surface temperature observation information is embodied in the analysis field;
in addition, compared with the traditional infrared bright temperature simulation scheme, the ground surface temperature and the atmospheric element profile of the input field are derived from the analysis field after the three-step assimilation, and the analysis field considers the constraint of the station ground surface temperature observation information on the logarithmic simulation field instead of the traditional numerical simulation field.
Drawings
FIG. 1 is a flow chart of a conventional infrared bright temperature simulation method;
FIG. 2 is a flow chart of an infrared bright temperature simulation method taking into account earth surface temperature observation information constraints;
FIG. 3 is a graph of the field effect of analysis of surface temperature after data assimilation for model forecast and station observation in an embodiment; in the figure, (a) is a numerical mode forecasting field; (b) actual observation for the station; (c) is an assimilation analytical field; (d) surface temperature increase for assimilation analysis;
FIG. 4 is a diagram showing the effect of Himaware-8/AHI infrared bright temperature simulation in the embodiment; in the figure, (a) is the simulation of the brightness temperature by the traditional method; (b) simulating the bright temperature for the method of the embodiment of the invention; (c) is the bright temperature observed by the AHI; (d) Simulating the difference of the brightness temperature for the actual observation and the traditional method; (e) Simulating the difference of the brightness temperature for the actual observation and the method of the embodiment of the invention;
FIG. 5 is a box diagram of the characteristics of the day change of the difference (unit: K) between the light temperature and the observed light temperature of the AHI channel 11 (8.6 μm) simulated by the conventional infrared light temperature simulation method (gray) and the infrared light temperature simulation method (white) in the embodiment of the invention; in the figure, (a) the statistical period is 6-8 months in 2018; (b) statistical period of from 12 months 2018 to 2 months 2019.
Detailed Description
The technical scheme of the invention is described in detail below through the drawings, but the protection scope of the invention is not limited to the embodiments.
Examples: a static satellite land infrared bright temperature simulation method constrained by earth surface temperature observation information. In this embodiment, the specific steps of the method will be described in detail using a Himaware-8 stationary satellite AHI infrared imager as an example.
S1, acquiring data of numerical weather mode forecast and ground surface temperature data of live observation of a reference station, and respectively preprocessing the two types of data; in the aspect of station ground surface temperature data, the live observation data of the national ground meteorological station are preprocessed, and parameters such as longitude, latitude, station altitude, ground surface temperature and the like are extracted. In the aspect of numerical mode data, preprocessing is carried out on simulation data of a WRF numerical weather forecast mode, wherein the preprocessing comprises the steps of extracting parameters such as the temperature of the atmosphere, water vapor and air pressure, the surface temperature of the ground, a wind field and the like.
S2, quality control is carried out on the numerical mode data and the surface temperature data which are preprocessed in the step S1.
In order to ensure reliable quality of the station surface temperature data participating in the assimilation algorithm, quality control is required to be performed on the data, a quality control scheme of the national grade ground weather station surface temperature data is established, and preparation is made for the assimilation algorithm of the numerical mode data and the station surface temperature data. The specific contents include: significant error checking, background field outlier checking, spatial consistency checking, temporal consistency checking, and station and pattern terrain height consistency checking.
(1) Major error checking
Through sample statistics, the earth surface temperature data of the meteorological station in China is found(superscript o indicates observation) the normal value range is between-50 and 70 ℃. Thus, a significant error checking threshold γ is set min ＝-50，γ max ＝70。
If it isThen reject +.>Data;
(2) Background field outlier verification
First, the numerical values areSurface temperature of mode simulation (Skin Surface Temperature)(superscript b indicates the mode, i and j indicate the horizontal coordinate number of the mode), performing spatial linear interpolation, interpolating to the spatial position of the ground weather station data, and obtaining the mode surface temperature +.>
Second, estimateObservation errors of ground surface temperature data are counted according to long-time samples, a triangle cap method is adopted, and ground surface temperature data inverted by MODIS satellites are utilized +.>And surface temperature data of GFS analytical field +.>Interpolation to the observation station point position, and calculation of the observation error sigma of the station surface temperature STA ，
wherein ,
again, an outlier check is performed:
if it isThen reject +.>Data;
(3) Spatial consistency check
First, each site is calculated separately using a horizontal distance of 220km (about 2 ° latitude and longitude) as a radius for spatial consistency checkMean +.about.within 220km radius from its surroundings>The difference between them, noted as:
secondly, a spatial consistency check is performed:
if it isThen reject +.>Data;
(4) Time consistency check
First, each site is calculated separately using one hour as a time window for time consistency checkAverage value +.about.1 hour before and after its time>The difference between them, noted as:
secondly, a time consistency check is performed:
if it isThen reject +.>Data;
(5) Station and mode terrain height consistency test
First, the topography height H of the numerical mode is interpolated in a bilinear manner b (i, j) (superscript b indicates a pattern), interpolating to the spatial position of the ground weather station data to obtain a pattern terrain height H at the nth ground weather station position b (n) and calculating the station height H o (n) the difference from the pattern terrain height |H o (n)-H m (n)|。
Second, a high consistency check is performed:
if |H o (n)-H m (n) | > 50m, then cull the nth siteData.
S3, establishing a data assimilation algorithm of WRF numerical mode data and station ground surface temperature live observation data. Reading numerical mode analog dataAnd S2, station surface temperature observation data after quality control>The algorithm adopts variation analysis as a mathematical method and is built based on the traditional three-dimensional variation functional. The method comprises the following steps:
the method for assimilating the two types of data in the step S3 comprises the following steps:
s31, increasing the observation of the surface temperature of the station in the variation assimilation cost function of the formula (7)To establish a variation assimilation cost function taking into account the observation of the station surface temperature, as in equation (8),
J(x)＝J B +J O (7)；
J(x)＝J B +J O +J ts (8)；
wherein, a background field penalty term is includedObservation penalty term other than surface temperature +.>And a newly added surface temperature observation penalty termEach penalty term x represents a value comprising the surface temperature analysis value +.>Is the assimilation analytical field, x b Representing the inclusion of surface temperature forecast values->The model background field of (a), H represents an observation operator, O represents an observation error matrix, and B represents a background error covariance matrix;
s32, improving the background error covariance matrix B to B new The background field penalty term becomes
B new To increase the surface temperature T in the original background error covariance B s And establish the surface temperature T s And air temperature T a Is related to B new As in (9),
wherein, psi, χ, T, RH, P s A flow function, a velocity potential function, an air temperature, a relative humidity, and an air pressure, respectively representing an atmospheric state quantity; χ_u, T a_u and Ps U represents the velocity potential function, temperature and air pressure of the unbalanced portion, respectively; t (T) s U is the unbalanced portion of the surface temperature; c (C) ψ,χ Representing the correlation between the velocity potential function χ and the flow function ψ;representing the temperature T a Correlation with the stream function ψ; />Representing the air pressure P s Correlation with the stream function ψ; i represents an identity matrix;
representing the surface temperature T s And air temperature T a Correlation between _u ++>In the form of a function of (c) is,
δT s (i,j)＝δT s _u(i,j)+δT s _b(i,j) (10)；
where i, j, k denote horizontal and vertical coordinates, b denote partitions of different features, δT s B is the equilibrium part of the surface temperature, and is represented by the air temperature T of formula (11) a Linear regression of u represents thePartially embody the surface temperature T s And air temperature T a Correlation of _u, and obtaining linear correlation coefficient of the two by sample statisticsAccording to the characteristics of regional differences of the surface temperature and the like, different subareas b are established at intervals of 5 degrees of latitude, and features under the condition of different latitudes are described;
realization of pairs by Cholesky decompositionInverting the matrix;
s33, adopting a classical conjugate gradient descent method to construct a gradient of the cost function newly constructed by the formula (8)Gradient decrease, when->Iterating the numerical values to the initial +.>1/1000 of the total number of the steps, converging the conjugate gradient descent iterative algorithm, and finally, the corresponding x is the optimal analysis field after functional minimization and comprises atmospheric elements and earth surface elements>Is a result of the analysis of (a).
S4, extracting atmospheric conventional elements (such as temperature, water vapor and air pressure) and earth surface elements (earth surface temperature and the like) corresponding to each horizontal space grid point, and adopting a UW_HSRemis earth surface emissivity dataset changing month by month as an input of a radiation transmission model CRTM, wherein the data set is used for simulating the bright temperatures of 10 infrared channels of an AHI instrument; and converting the simulated bright temperature value corresponding to the horizontal space grid point into an AHI pixel point. Compared with the traditional infrared bright temperature simulation scheme, the improvement is that: the surface temperature and atmospheric element profile of the input field are derived from the analysis field after the step three assimilation, which takes into account the constraints of the station surface temperature observation information on the numerical simulation field instead of the traditional numerical simulation field.
Specific analysis of the method of the examples is carried out with reference to the accompanying drawings:
fig. 1 is a diagram of a conventional scheme, in which atmospheric normal elements (temperature, humidity, air pressure) and surface elements (surface temperature and air pressure) are extracted from numerical mode simulation data, each grid point is processed into a vertical profile and a single point form according to the sequence of the grid points, a static data set of surface emissivity is utilized in combination, a radiation transmission model is input, the satellite infrared light temperature on each grid point is calculated, and then the simulated light temperature corresponding to an AHI observation pixel point is obtained through interpolation. FIG. 2 is a view of the present invention, and is modified in that the numerical model forecast data is assimilated with the station surface temperature observation data in the part of the dashed box, so that the surface temperature and the atmospheric temperature analysis values are more realistic, and for each lattice point, the surface temperature single point information and the vertical profile of the atmospheric temperature are extracted from the assimilation analysis field, and the radiation transmission model is input by combining the use of the surface emissivity static data set, so as to calculate the infrared simulated bright temperature.
Fig. 3 shows distribution characteristics of the surface temperature data in numerical mode, the station surface temperature observation data and the surface temperature observation analysis value after data assimilation, and shows the effect of adding the constraint of the surface temperature observation information. From the effect of the example of 24 days (UTC) of 8 months in 2018, the surface temperature of the model forecast is 3-5 ℃ lower than that of the live observation in the middle eastern region of China (Jiangsu, anhui, henan, shandong and other provinces), after the data is assimilated, the surface temperature assimilation analysis field of the graph (c) is closer to that of the graph (d), and after the data is assimilated, the influence of the graph (d) is shown as that the surface temperature of the middle eastern region is increased to a certain extent compared with that of the graph (a), and the adjustment is reasonable and correct through the comparison observation.
Fig. 4 shows a comparison between the infrared bright temperature simulated by the conventional scheme and the infrared bright temperature and the AHI observed bright temperature simulated by the new scheme. For 18 (UTC) cases in 2018 8, 8 and 24 months, taking AHI channel 11 (wavelength: 8.6 μm) as an example, the infrared bright temperature simulation of the traditional scheme in the middle eastern region of China is obviously lower than that observed by minus 2 to minus 4 ℃, as shown in a graph (d). The new scheme can well promote the simulation effect of the area, and is closer to the observed bright temperature, as shown in the figure (e).
Fig. 5 shows long-term sample statistics for month 6-8 in 2018 and month 12-2 in 2018, respectively. Compared with the traditional scheme, the bright temperature simulation deviation of the new scheme is obviously reduced (closer to 0), the deviation in the daytime period is obviously improved, and meanwhile, the standard deviation of bright temperature simulation is also obviously reduced, which indicates that the infrared bright temperature simulation scheme taking the table surface temperature observation information constraint into consideration has higher simulation precision.
As described above, although the present invention has been shown and described with reference to certain preferred embodiments, it is not to be construed as limiting the invention itself. Various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A static satellite land infrared bright temperature simulation method constrained by earth surface temperature observation information is characterized by comprising the following steps:
s1, acquiring data of numerical weather mode forecast and ground surface temperature data of live observation of a reference station, and respectively preprocessing the two types of data;
s2, quality control is carried out on the numerical mode data and the surface temperature data which are preprocessed in the step S1;
s3, assimilating the two types of data processed in the step S2, wherein the steps are as follows:
s31, increasing the observation of the surface temperature of the station in the variation assimilation cost function of the formula (1)To establish a variation assimilation cost function taking into account the observation of the station surface temperature, as in equation (2),
J(x)＝J B +J O (1)；
wherein, a background field penalty term is includedObservation penalty term other than surface temperatureAnd a newly added surface temperature observation penalty termEach penalty term x represents a value comprising the surface temperature analysis value +.>Is the assimilation analytical field, x b Representing the inclusion of surface temperature forecast values->The model background field of (a), H represents an observation operator, O represents an observation error matrix, and B represents a background error covariance matrix;
s32, improving the background error covariance matrix B to B new The background field penalty term becomes
B new To increase the surface temperature T in the original background error covariance B s And establish the surface temperature T s And air temperature T a Is related to B new As shown in (3),
wherein, psi, χ, T, RH, P s A flow function, a velocity potential function, an air temperature, a relative humidity, and an air pressure, respectively representing an atmospheric state quantity; χ_u, T a_u and Ps U represents the velocity potential function, temperature and air pressure of the unbalanced portion, respectively; t (T) s U is the unbalanced portion of the surface temperature; c (C) ψ,χ Representing the correlation between the velocity potential function χ and the flow function ψ;representing the temperature T a Correlation with the stream function ψ;representing the air pressure P s Correlation with the stream function ψ; i represents an identity matrix;
representing the surface temperature T s And air temperature T a Correlation between _u ++>In the form of a function of (c) is,
δT s (i,j)＝δT s _u(i,j)+δT s _b(i,j) (4)；
where i, j, k denote horizontal and vertical coordinates, b denote partitions of different features, δT s B is the equilibrium part of the surface temperature, and is represented by the air temperature T of formula (5) a Linear regression of u represents that the portion embodies the surface temperature T s And air temperature T a Correlation of _u, and obtaining linear correlation coefficient of the two by sample statisticsAccording to the characteristics of regional differences of the surface temperature and the like, different subareas b are established at intervals of 5 degrees of latitude, and features under the condition of different latitudes are described;
realization of pairs by Cholesky decompositionInverting the matrix;
s33, adopting a classical conjugate gradient descent method to carry out gradient on the newly constructed cost function of the formula (2)Gradient decrease, when->Iterating the numerical values to the initial +.>1/1000 of the total number of the steps, converging the conjugate gradient descent iterative algorithm, and finally, the corresponding x is the optimal analysis field after functional minimization and comprises atmospheric elements and earth surface elements>Is determined by the analysis value of (a);
s4, extracting the atmospheric elements and the earth surface elements corresponding to each horizontal space grid point obtained in the step S3, adopting an earth surface emissivity data set changing month by month as the input of a radiation transmission model, simulating the bright temperature of an infrared channel of the static satellite infrared imager, and then converting the simulated bright temperature value corresponding to the horizontal space grid point into the pixel point of the static satellite infrared imager to complete the static satellite terrestrial infrared bright temperature simulation.
2. The method for simulating the infrared bright temperature of the stationary satellite on land constrained by the surface temperature observation information according to claim 1, wherein the preprocessing of the data for the log weather pattern forecast in the step S1 comprises: extracting longitude, latitude, station altitude and surface temperature;
preprocessing of ground surface temperature data live observed by a reference station includes extracting the temperature of the atmosphere, water vapor and air pressure, and ground surface temperature and wind field.
3. The method for simulating the infrared bright temperature of the stationary satellite constrained by the surface temperature observation information according to claim 1, wherein the specific content of performing quality control on the two types of data in the step S2 comprises the following steps: significant error checking, background field outlier checking, spatial consistency checking, temporal consistency checking, and station and pattern terrain height consistency checking.
4. A method of simulating a stationary satellite terrestrial infrared bright temperature constrained by surface temperature observation information according to claim 3, wherein the step of significant error checking is:
setting a critical error checking threshold gamma min ＝-50，γ max ＝70；
If it isThen reject +.>Data,/->Is the earth surface temperature observation data of the nth space station.
5. The method for simulating the infrared bright temperature of the stationary satellite constrained by the surface temperature observation information according to claim 3, wherein the step of background field outlier verification is as follows:
first, the surface temperature of the numerical mode simulation is simulatedPerforming spatial linear interpolation, wherein the superscript b represents a mode, i and j represent horizontal coordinate numbers of the mode, interpolating to spatial positions of ground weather station data, and obtaining the mode surface temperature +.>
Second, estimateObservation errors of ground surface temperature data are counted according to long-time samples, a triangle cap method is adopted, and ground surface temperature data inverted by MODIS satellites are utilized +.>And surface temperature data of GFS analytical field +.>Interpolation to the observation station point position, and calculation of the observation error sigma of the station surface temperature STA ，
wherein ,
then performing outlier verification:
if it isThen reject +.>Data.
6. A method of simulating a stationary satellite terrestrial infrared bright temperature constrained by surface temperature observation information according to claim 3, wherein the step of spatial consistency checking comprises:
first, the radius of the space consistency test is calculated for each site with a horizontal distance of 220kmMean +.about.within 220km radius from its surroundings>The difference between them, noted as:
then a spatial consistency check is performed:
if it isThen reject +.>Data.
7. The method for simulating the infrared bright temperature of the stationary satellite constrained by the surface temperature observation information according to claim 3, wherein the specific steps of the time consistency test are as follows:
first, each site is calculated separately using one hour as a time window for time consistency checkAverage value +.about.1 hour before and after its time>The difference between them, noted as:
then, a time consistency check is performed:
if it isThen reject +.>Data.
8. The method for simulating the infrared bright temperature of the stationary satellite constrained by the surface temperature observation information according to claim 3, wherein the specific steps of the test station and pattern terrain height consistency test are as follows:
first, the topography height H of the numerical mode is interpolated in a bilinear manner b (i, j) interpolating to the spatial position of the ground weather station data to obtain the model terrain height H at the nth ground weather station position b (n) and calculating the station height H o (n) the difference from the pattern terrain height |H o (n)-H m (n)|；
A high consistency check is then performed:
if |H o (n)-H m (n)|>50m, reject the nth siteData.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210892133.9A CN116341352B (en) | 2022-07-27 | 2022-07-27 | Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210892133.9A CN116341352B (en) | 2022-07-27 | 2022-07-27 | Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116341352A CN116341352A (en) | 2023-06-27 |
| CN116341352B true CN116341352B (en) | 2023-10-24 |
Family
ID=86891788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210892133.9A Active CN116341352B (en) | 2022-07-27 | 2022-07-27 | Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116341352B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580003A (en) * | 2018-12-18 | 2019-04-05 | 成都信息工程大学 | A kind of stationary weather satellite Thermal Infrared Data estimation Air Close To The Earth Surface temperature methods |
| CN110749942A (en) * | 2019-11-12 | 2020-02-04 | 江苏省气象科学研究所 | Clear sky channel detection quality control method suitable for data assimilation of stationary satellite |
| CN112834040A (en) * | 2020-12-24 | 2021-05-25 | 合肥冬行信息科技有限公司 | Ground thermal anomaly identification algorithm based on stationary satellite |
| CN113111529A (en) * | 2021-04-22 | 2021-07-13 | 南京气象科技创新研究院 | Infrared brightness temperature simulation method fusing numerical value mode and satellite microwave cloud inversion data |
| CN113189678A (en) * | 2021-04-29 | 2021-07-30 | 南京气象科技创新研究院 | Rainfall phenomenon comprehensive identification method and device based on multi-source observation data |
-
2022
- 2022-07-27 CN CN202210892133.9A patent/CN116341352B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580003A (en) * | 2018-12-18 | 2019-04-05 | 成都信息工程大学 | A kind of stationary weather satellite Thermal Infrared Data estimation Air Close To The Earth Surface temperature methods |
| CN110749942A (en) * | 2019-11-12 | 2020-02-04 | 江苏省气象科学研究所 | Clear sky channel detection quality control method suitable for data assimilation of stationary satellite |
| CN112834040A (en) * | 2020-12-24 | 2021-05-25 | 合肥冬行信息科技有限公司 | Ground thermal anomaly identification algorithm based on stationary satellite |
| CN113111529A (en) * | 2021-04-22 | 2021-07-13 | 南京气象科技创新研究院 | Infrared brightness temperature simulation method fusing numerical value mode and satellite microwave cloud inversion data |
| CN113189678A (en) * | 2021-04-29 | 2021-07-30 | 南京气象科技创新研究院 | Rainfall phenomenon comprehensive identification method and device based on multi-source observation data |
Non-Patent Citations (2)
| Title |
|---|
| 地基微波辐射计亮温观测数据的质量评价、订正与反演应用研究;李青;《中国博士全文数据库 基础科学辑》(第1期);1-120 * |
| 晨昏轨道微波温度计资料同化对降水定量预报的影响及其对三轨卫星系统的;邹晓蕾 等;《大气科学》;第40卷(第1期);46-62 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116341352A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109580003B (en) | Method for estimating near-ground atmospheric temperature by thermal infrared data of stationary meteorological satellite | |
| Boilley et al. | Comparison between meteorological re-analyses from ERA-Interim and MERRA and measurements of daily solar irradiation at surface | |
| Adler et al. | Tropical rainfall distributions determined using TRMM combined with other satellite and rain gauge information | |
| Doherty et al. | A gridded reconstruction of land and ocean precipitation for the extended tropics from 1974 to 1994 | |
| CN110927120B (en) | Early warning method for vegetation coverage | |
| Zhang et al. | Development of the direct-estimation albedo algorithm for snow-free Landsat TM albedo retrievals using field flux measurements | |
| Lim et al. | A land data assimilation system using the MODIS-derived land data and its application to numerical weather prediction in East Asia | |
| Ouarda et al. | The generalized additive model for the assessment of the direct, diffuse, and global solar irradiances using SEVIRI images, with application to the UAE | |
| Tang et al. | Dependence of remote sensing accuracy of global horizontal irradiance at different scales on satellite sampling frequency | |
| CN113111529B (en) | Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data | |
| Li et al. | A global zenith tropospheric delay model with ERA5 and GNSS-based ZTD difference correction | |
| Wang et al. | Estimation of surface longwave radiation over the Tibetan plateau region using MODIS data for cloud-free skies | |
| CN113408111B (en) | Atmospheric precipitation inversion method and system, electronic equipment and storage medium | |
| Li et al. | An Operational Split-Window Algorithm for Generating Long-Term Land Surface Temperature Products from Chinese Fengyun-3 Series Satellite Data | |
| Grant et al. | Meteorological and remotely sensed datasets for hydrological modelling: A contribution to the Australian Water Availability Project | |
| CN116341352B (en) | Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint | |
| Fan et al. | Extension of the generalized split-window algorithm for land surface temperature retrieval to atmospheres with heavy dust aerosol loading | |
| Huang et al. | A novel global grid model for atmospheric weighted mean temperature in real-time GNSS precipitable water vapor sounding | |
| Tian et al. | Improving MODIS aerosol estimates over land with the surface BRDF reflectances using the 3-D discrete cosine transform and RossThick-LiSparse models | |
| Jedlovec et al. | A satellite-derived upper-tropospheric water vapor transport index for climate studies | |
| Shahabadi et al. | Impact of slant-path radiative transfer in the simulation and assimilation of satellite radiances in Environment Canada’s weather forecast system | |
| CN115541023A (en) | Land surface temperature and land surface emissivity synchronous simulation inversion method and system | |
| Zhao et al. | An operational land surface temperature retrieval methodology for Chinese second-generation Huanjing disaster monitoring satellite data | |
| CN113255148A (en) | Method for estimating all-weather air temperature and space-time distribution thereof based on MODIS product data | |
| Wang et al. | Spatial distribution modeling of temperature increase for the uplifted mountain terrains and its characteristics in Southwest 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |