CN113111529A - Infrared brightness temperature simulation method fusing numerical value mode and satellite microwave cloud inversion data - Google Patents

Infrared brightness temperature simulation method fusing numerical value mode and satellite microwave cloud inversion data Download PDF

Info

Publication number
CN113111529A
CN113111529A CN202110436628.6A CN202110436628A CN113111529A CN 113111529 A CN113111529 A CN 113111529A CN 202110436628 A CN202110436628 A CN 202110436628A CN 113111529 A CN113111529 A CN 113111529A
Authority
CN
China
Prior art keywords
cloud
data
satellite
microwave
atms
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110436628.6A
Other languages
Chinese (zh)
Other versions
CN113111529B (en
Inventor
李昕
曾明剑
汪宁
唐飞
诸葛小勇
邹晓蕾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Institute Of Meteorological Science And Technology Innovation
Original Assignee
Nanjing Institute Of Meteorological Science And Technology Innovation
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing Institute Of Meteorological Science And Technology Innovation filed Critical Nanjing Institute Of Meteorological Science And Technology Innovation
Priority to CN202110436628.6A priority Critical patent/CN113111529B/en
Publication of CN113111529A publication Critical patent/CN113111529A/en
Application granted granted Critical
Publication of CN113111529B publication Critical patent/CN113111529B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • 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
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08—Thermal analysis or thermal optimisation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses an infrared brightness temperature simulation method fusing a numerical mode and satellite microwave cloud inversion data, and relates to the field of atmospheric science research, wherein the method specifically comprises the following steps: 1) preprocessing WRF numerical simulation data and satellite ATMS microwave cloud inversion data; 2) performing quality control on satellite ATMS microwave cloud inversion data; 3) and carrying out data fusion on the WRF numerical simulation data and the satellite ATMS microwave cloud inversion data after quality control to obtain a fusion analysis field of atmospheric cloud water, cloud ice and rainwater. 4) And (3) extracting atmospheric conventional elements of numerical simulation data and atmospheric cloud water substance elements of fusion analysis data to form vertical profiles, using the vertical profiles as input profiles of a radiation transmission model CRTM, calculating satellite infrared simulation brightness temperature on each grid point, and converting the satellite infrared simulation brightness temperature value into a CrIS view field by adopting a CrIS observation and grid point data view field matching technology. The invention improves the simulation level of the cloud area.

Description

Infrared brightness temperature simulation method fusing numerical value mode and satellite microwave cloud inversion data
Technical Field
The invention relates to the field of atmospheric science research, in particular to an infrared brightness temperature simulation method for a CrIS infrared hyperspectral exploration channel of a Suomi-NPP polar orbit satellite, which integrates a numerical mode and satellite microwave cloud inversion data.
Background
With the continuous development of the meteorological polar orbit satellite observation technology, the data of the infrared detector and the hyperspectral detector are widely applied to the fields of weather monitoring, weather early warning, weather analysis and numerical weather forecast. Compared with the traditional infrared detector, the infrared hyperspectral detector has a large number of channels, and the channels with different frequencies are sensitive to different height layers of the earth surface or the atmosphere, so that the infrared hyperspectral detector becomes an important observation information source in numerical weather forecast. On one hand, the numerical mode requires the satellite infrared bright temperature observation for data assimilation; on the other hand, the numerical mode also requires the satellite infrared light temperature observation for forecast inspection. In order to achieve the above purpose, the channel brightness and temperature of the satellite infrared instrument are simulated by the atmospheric factors, and this process is called infrared brightness and temperature simulation.
As shown in fig. 1a, as a conventional infrared radiation simulation method, a satellite infrared bright temperature simulation adopts a mode that atmospheric temperature, water vapor, air pressure, cloud water information, surface information, geometric parameters corresponding to a satellite field of view and the like are input into a radiation transmission model, absorption and scattering effects of absorption gas and cloud in the atmosphere on radiation, surface radiation effects, solar short wave radiation effects and the like are calculated, radiation entering a satellite sensor from the top of an atmospheric layer is simulated by a radiation transmission equation, and then the radiation is converted into bright temperature. Since infrared radiation is sensitive to clouds, the performance of the bright-temperature simulation depends greatly on the description of the radiation transmission model on the absorption and scattering effects of the clouds, on one hand, the performance depends on the parameterization of extinction coefficients, single-shot reflectance, scattering phase functions and the like for different clouds in the radiation transmission model, and on the other hand, the performance also depends strongly on the cloud profile structure of the input radiation transmission model.
Reviewing the existing infrared radiation simulation method, researchers often adopt rapid radiation modes such as CRTM and RTTOV as models, and establish the infrared radiation simulation method by taking atmospheric elements forecasted by a numerical weather mode as model input quantity. The DingweiYu and Wanqilin (2008) uses WRF numerical mode data as an input field, and utilizes RTTOV to simulate the Infrared channel brightness temperature of High Resolution infracted Radiation Sounder (HIRS/3); guo xing et al (2016) uses the WRF numerical model prediction data as an input field, and uses CRTM to simulate the channel brightness temperature of an Atmospheric Infrared Sounder (AIRS) hyperspectral Atmospheric Infrared detector; okamoto et al (2017) simulated the channel brightness temperature of an Advanced Himapwari Imager (AHI) by CRTM using JMA-NHM mesoscale numerical mode data as an input field. The limitations of these methods are that there is a great uncertainty about the cloud rain distribution and vertical structure of the numerical weather pattern simulation (Faijan et al 2012; Li et al 2016), and the cloud absorption and scattering effects calculated by the radiation transmission model have a great error, which results in an undesirable infrared radiation simulation effect, especially in the cloud region. In view of this, the satellite infrared brightness temperature simulation method needs to be further improved.
As described above, one of the ways to improve the satellite infrared brightness and temperature simulation is to improve the profile information of cloud water, cloud ice, rainwater, etc. input into the radiation transmission model, and the cloud water structure of the numerical model can be modified in combination with other observation information in the existing method. Polar orbit meteorological satellites are generally equipped with infrared detection instruments and microwave detection instruments at the same time. Microwaves have better penetration for clouds than infrared and can be used for inverting atmospheric cloud water vertical structures, for example, a linear regression method of window channel (Weng and Grody 1994; Weng et al 2003) is used for inverting a cloud water path and a cloud ice path, a one-dimensional variational method (e.g., Microwave Integrated recovery System, MiRS System) is used for inverting a cloud water, cloud ice and rainwater profile (Boukabara et al 2011) and the like, so Microwave detection can be used as supplementary information for infrared light temperature simulation. For example, in a commonly used meteorological polar orbit satellite, a Cross-track isolated Sounder (CrIS) Infrared hyperspectral detector and an Advanced Technology Microwave detector (ATMS) are mounted on an American Suomi-NPP satellite which is emitted in 11 months 2011, and ATMS Microwave cloud inversion data can provide support for the simulation of the Infrared bright temperature of CrIS.
The existing satellite infrared brightness temperature simulation method adopts simulation data of a numerical mode as an input profile, when a cloud area exists, the numerical mode has large simulation errors on the structure and distribution of cloud water, cloud ice and rainwater, and if the numerical mode data is directly applied without correction, the cloud effect description errors in radiation transmission simulation are large, so that the satellite infrared brightness temperature simulation effect is not ideal.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide an infrared brightness temperature simulation method fusing a numerical mode and satellite microwave cloud inversion data, which is suitable for infrared brightness temperature simulation of a Suomi-NPP polar orbit satellite CrIS infrared hyperspectral detector, can improve the infrared brightness temperature simulation effect of CrIS in a cloud area, and improves the application capability of CrIS in a numerical prediction mode.
The technical method comprises the following steps: in consideration of the accuracy problem of the existing satellite infrared brightness temperature simulation method in a cloud area, the method provided by the invention utilizes more accurate microwave observation information to improve the accuracy of common numerical weather mode data, develops a fusion algorithm of corresponding numerical mode data and satellite microwave inversion data, and establishes an improved satellite infrared brightness temperature simulation method. An improved infrared bright temperature simulation method is established for a CrIS infrared hyperspectral detector on a polar orbit satellite Suomi-NPP. The method is to combine ATMS microwave cloud inversion data and Weather Research and Forecasting (WRF) numerical mode data on a Suomi-NPP satellite, realize the fusion of the satellite microwave cloud inversion data and the numerical mode prediction data by establishing a corresponding data fusion process flow, further use an optimized cloud water substance fusion field as an input field of a radiation transmission model, establish an infrared bright temperature simulation method, and improve the rationality and accuracy of simulation.
An infrared brightness temperature simulation method fusing a numerical mode and satellite microwave cloud inversion data comprises the following steps:
preprocessing WRF numerical simulation data and satellite ATMS microwave cloud inversion data;
preprocessing the WRF numerical simulation data, namely preprocessing the simulation data of the WRF numerical weather forecast mode, and extracting conventional atmospheric elements, surface elements and atmospheric cloud water substance elements;
preprocessing the satellite ATMS microwave cloud inversion data, namely preprocessing the ATMS microwave cloud inversion data generated by the MiRS system, and extracting atmospheric cloud water substance elements;
step two, performing quality control on satellite micro ATMS microwave cloud inversion data;
performing data fusion on WRF numerical simulation data and satellite ATMS microwave cloud inversion data after quality control to obtain a fusion analysis field of atmospheric cloud and water substance elements;
extracting atmospheric conventional elements of the WRF numerical simulation data and atmospheric cloud water substances of the fusion analysis field as vertical profiles, and taking the vertical profiles as input profiles of the radiation transmission model CRTM; extracting earth surface elements of the WRF numerical simulation data as single-point elements, inputting the single-point elements and geometric parameters corresponding to the satellite field of view into a radiation transmission model CRTM, and calculating satellite infrared simulation brightness temperature on each grid point; and converting the infrared simulated brightness temperature value of the satellite into a CrIS view field by adopting a CrIS observation and grid data view field matching technology, calculating the weight of each grid simulated value in a CrIS view field range through a two-dimensional Gaussian distribution function, and further obtaining the average simulated brightness temperature through weighted average to serve as the final CrIS view field simulated brightness temperature.
Further, the atmospheric conventional elements include temperature, water vapor and air pressure; the surface elements comprise surface temperature and surface air pressure; the atmospheric cloud water substance elements comprise a cloud water mixing ratio, a cloud ice mixing ratio and a rainwater mixing ratio.
Further, the method for controlling the quality of the satellite micro ATMS microwave cloud inversion data comprises the steps of significant error detection, background field outlier detection and space consistency detection;
1) and (3) major error detection:
cloud-water mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000031
Integrating in vertical direction to obtain the path of cloud water
Figure BDA0003033378120000041
Wherein the subscript c represents the cloud water, and the superscript o represents the observation, which is denoted as the actual integration, CVI for short;
cloud ice mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA00030333781200000410
Integrating in vertical direction to obtain the path of ice cloud
Figure BDA0003033378120000042
Wherein the subscript i represents cloud ice, and the superscript o represents observation, which is denoted as actual integration, abbreviated as IVI;
rainwater mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000043
Integrating in the vertical direction to obtain the rain path
Figure BDA0003033378120000044
Wherein the subscript r represents Rain water, and the superscript o represents observation, which is marked as Rain vertical integration, RVI for short;
in the three formulas, i represents the track crossing direction, j represents the track direction, k is vertical layering, rho represents the atmospheric density, z represents the vertical height, and kmax represents the vertical serial number corresponding to the maximum height value;
the normal value range of CVI, IVI and RVI is found to be 0-3000g m through sample statistics-2And thus, a critical error checking threshold gamma is setgross＝3000g·m-2；
If CVIo(i,j)＞γgross,or,CVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200000411
Data;
if IVIo(i,j)＞γgross,or,IVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200000412
Data;
if RVIo(i,j)＞γgross,or,RVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200000413
Data;
2) background field outlier verification test:
firstly, the cloud-water mixing ratio of WRF numerical mode analog data
Figure BDA0003033378120000045
Cloud ice mixing ratio
Figure BDA0003033378120000046
Rainwater mixing ratio
Figure BDA0003033378120000047
Performing space linear interpolation to obtain satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000048
And
Figure BDA0003033378120000049
wherein subscript c represents cloud water, subscript i represents cloud ice, subscript r represents rain water, and superscript m represents simulation;
secondly, vertical integration is carried out to obtain the cloud water path CVI of the WRF numerical simulation datam(i, j), cloud ice pathway IVIm(i, j) and rainwater Path RVIm(i, j); calculating standard deviation sigma of cloud water path, cloud ice path and rainwater path according to the statistical sampleCVI、σCVIAnd σCVIWherein, in the step (A),
Figure BDA0003033378120000051
Figure BDA0003033378120000052
Figure BDA0003033378120000053
again, an outlier check is performed:
if CVIo(i,j)-CVIm(i,j)|＞3·σCVIThen eliminate the space point (i, j), all height layers
Figure BDA00030333781200000512
Data;
if IVIo(i,j)-IVIm(i,j)|＞3·σIVIThen eliminate the space point (i, j), all height layers
Figure BDA00030333781200000513
Data;
if RVIo(i,j)-RVIm(i,j)|＞3·σRVIThen eliminate the space point (i, j), all height layers
Figure BDA00030333781200000514
Data;
3) and (3) checking the spatial consistency:
first, the CVI of each spatial point (i, j) is calculated separately with the horizontal distance of 110km as the radius of the spatial consistency checko(i,j)、IVIo(i, j) and RVIo(i, j) and the average value within a radius of 110km around (i, j)
Figure BDA0003033378120000054
Figure BDA0003033378120000055
And
Figure BDA0003033378120000056
the differences between, respectively noted:
Figure BDA0003033378120000057
Figure BDA0003033378120000058
Figure BDA0003033378120000059
second, a spatial consistency check is performed:
if deltaCVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200000515
Data;
if deltaIVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200000516
Data;
if deltaRVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200000517
And (4) data.
The fusion algorithm for performing data fusion on the WRF numerical simulation data and the satellite ATMS microwave cloud inversion data after quality control in the third step is specifically as follows:
reading WRF numerical mode analog data
Figure BDA00030333781200000510
And
Figure BDA00030333781200000511
satellite ATMS microwave inversion data
Figure BDA0003033378120000061
And
Figure BDA0003033378120000062
and converting into a one-dimensional vector; using variational analysis as a mathematical method of an optimized fusion algorithm, constructing a functional by using cloud water, cloud ice and rainwater data which are subjected to WRF numerical simulation and cloud water, cloud ice and rainwater data which are subjected to satellite ATMS microwave inversion, and solving an extreme value of the functional by using a conjugate gradient descent method so as to solve an optimal fusion analysis field of the cloud water, the cloud ice and the rainwater;
the specific algorithm is as follows:
definition of functional Jc、JiAnd Jr，
Figure BDA0003033378120000063
Figure BDA0003033378120000064
Figure BDA0003033378120000065
Wherein three-dimensional variables are combined
Figure BDA0003033378120000066
Is recorded as a one-dimensional vector form
Figure BDA0003033378120000067
To make three-dimensional variable
Figure BDA0003033378120000068
Is recorded as a one-dimensional vector form
Figure BDA0003033378120000069
To make three-dimensional variable
Figure BDA00030333781200000610
Is recorded as a one-dimensional vector form
Figure BDA00030333781200000611
Wherein the content of the first and second substances,
Figure BDA00030333781200000612
and
Figure BDA00030333781200000613
has a dimension of nx1×ny1×nz1，nx1、ny1And nz1Respectively representing the total number of lattice points of the WRF numerical mode simulation data in the east-west direction, the south-north direction and the vertical direction;
to make three-dimensional variable
Figure BDA00030333781200000614
Is recorded as a one-dimensional vector form
Figure BDA00030333781200000615
To make three-dimensional variable
Figure BDA00030333781200000616
Is recorded as a one-dimensional vector form
Figure BDA00030333781200000617
To make three-dimensional variable
Figure BDA00030333781200000618
Is recorded as a one-dimensional vector form
Figure BDA00030333781200000619
Wherein the content of the first and second substances,
Figure BDA00030333781200000620
and
Figure BDA00030333781200000621
has a dimension of nx2×ny2×nz2，nx2、ny2And nz2Respectively represent the ATMS microwave inversion data of the satelliteThe total number of grid points in the cross-track direction, along-track direction and vertical direction;
is prepared from flos Chrysanthemi IndicicError covariance matrix recorded as numerical mode cloud and water mixing ratio data
Figure BDA00030333781200000622
ΜiError covariance matrix recorded as numerical mode cloud ice mixing ratio data
Figure BDA00030333781200000623
ΜrError covariance matrix recorded as numerical mode rainwater mixing ratio data
Figure BDA00030333781200000624
Wherein the content of the first and second substances,
Figure BDA00030333781200000625
and
Figure BDA00030333781200000626
has a dimension of nx1×ny1×nz1；Μc、ΜiAnd μmrHas a dimension of (n)x1×ny1×nz1)2；
Mixing O withcError covariance matrix recorded as satellite microwave cloud and water mixing ratio data
Figure BDA00030333781200000627
OiError covariance matrix recorded as satellite microwave cloud ice mixing ratio data
Figure BDA00030333781200000628
OrError covariance matrix recorded as satellite microwave rainwater mixing ratio data
Figure BDA00030333781200000629
Wherein the content of the first and second substances,
Figure BDA00030333781200000630
and
Figure BDA00030333781200000631
has a dimension of nx2×ny2×nz2；Oc、OiAnd OrHas a dimension of (n)x2×ny2×nz2)2；
In the formulas (7) - (9), H is an interpolation operator for converting the numerical mode data space into the satellite microwave data space, and superscript T represents matrix transposition;
the core solving process of the fusion algorithm is to Jc、JiAnd JrMinimization of the functional to obtain Qc、QiAnd QrThe fusion analysis field is the optimal fusion analysis field.
Further, the solution of the functional minimum value of the cloud-water mixing ratio is as follows:
the following equivalence relations are established,
ΔQ＝Q-Qm (10)
d＝H(Qm)-Qo (11)
then equation (7) can be rewritten as an incremental form with the subscript c hidden, which is equivalent to,
Figure BDA0003033378120000071
where H is the first order differential form of H, taking into account the first order Taylor's expansion approximation, i.e.
Figure BDA0003033378120000073
Decomposing the error covariance matrix (Md) of the pattern data into a horizontal covariance matrix (D)hAnd vertical covariance matrix DvI.e.. mu. Dv TDh TDhDvThen, the formula (12) is equivalent to,
Figure BDA0003033378120000072
the gradient of the equation (13) to the delta Q is obtained,
▽J(ΔQ)＝(DhDv)-1ΔQ+HTO-1(HΔQ-d) (14)
before solving the gradient descent algorithm, preprocessing is carried out, x is recorded as,
x＝(DhDv)-1ΔQ (15)
then equation (14) is equivalent to:
▽J(x)＝x+HTO-1(HDhDvx-d) (16)
solving gradient descent for ^ J (x) by adopting a classical conjugate gradient descent method, converging a conjugate gradient descent iteration algorithm when a value of ^ J (x) is iterated to 1/1000 of the initial ^ J (x), wherein x corresponding to convergence is a vector corresponding to the minimization of a functional;
finally, Δ Q can be calculated from x by equation (15), and Q can be calculated from Δ Q by equation (10), where Q is a one-dimensional vector corresponding to the final fusion analysis field, and transposing Q into a three-dimensional field is the final result of the fusion algorithm, i.e., the fusion analysis field, because Q is the same as Qc、QiAnd QrThe variational analysis algorithms of (a) are the same and are therefore denoted collectively by the symbol Q.
Further, the weight expression of the corresponding grid point calculated by the two-dimensional gaussian function in the fourth step is as follows:
Figure BDA0003033378120000081
wherein x and y represent the horizontal coordinates of the grid field, x0And y0Represents the horizontal coordinate corresponding to the center of CrIS field of view, LxAnd LyThe characteristic scale is represented, and w (x, y) represents the weight corresponding to the (x, y) grid point.
Has the advantages that: 1) the invention improves the infrared brightness temperature simulation level of the polar orbit satellite infrared hyperspectral data, in particular to the simulation level in a cloud area; 2) compared with the traditional infrared bright temperature simulation method, the method has the advantages that firstly, the cloud water substance elements simulated in the numerical mode and cloud water substance element data inverted by satellite microwaves are subjected to fusion analysis, the vertical profile of the cloud water substance elements is extracted from a fusion analysis field, then the vertical profile is input into a radiation transmission model, the infrared simulated bright temperature is calculated, the simulation error is reduced, the infrared simulated bright temperature simulation is more reasonable, and the accuracy is higher; and secondly, field matching is carried out on the CrIS observation field and the lattice data, the weight of each lattice analog value in the CrIS field range is calculated through a two-dimensional Gaussian function, the final CrIS field simulation bright temperature is calculated through weighted average, and the accuracy of the infrared simulation bright temperature is further improved.
Drawings
Fig. 1a is a flow chart of a conventional infrared brightness temperature simulation method.
FIG. 1b is a flow chart of an infrared brightness temperature simulation method for fusing a numerical model and satellite microwave cloud inversion data.
Fig. 2 is a field effect comparison graph of cloud water, cloud ice and rain water analysis of numerical model simulation data, satellite microwave inversion data and fusion algorithm analysis data.
Fig. 2a is a CVI distribution diagram of a cloud water mixing ratio path processed by numerical mode simulation data.
Fig. 2b is a cloud ice mixing ratio path IVI profile processed using numerical mode simulation data.
Fig. 2c is a rain mixture ratio path RVI profile processed using numerical mode simulation data.
FIG. 2d is a diagram of a cloud water mixing ratio path CVI distribution processed using satellite microwave inversion data.
Fig. 2e is a cloud ice mixing ratio path IVI profile processed using satellite microwave inversion data.
Fig. 2f is a stormwater mixing ratio path RVI profile processed using satellite microwave inversion data.
Fig. 2g is a cloud water mix ratio path CVI profile processed with fused data.
Fig. 2h is a cloud ice mix ratio path IVI profile processed with fused data.
Fig. 2i is a stormwater mixing ratio path RVI profile processed with fused data.
FIG. 3a is a schematic diagram of CrIS subsatellite observation field matching with lattice field.
FIG. 3b is a Gaussian weight for each bin analog value within the CrIS field of view.
Fig. 4a is a spatial distribution diagram for simulating the infrared light temperature by using a conventional infrared light temperature simulation method.
Fig. 4b is a spatial distribution diagram of simulated infrared light temperature using the method of the present invention.
FIG. 4c is a CrIS observed bright temperature spatial distribution plot.
FIG. 4d is CrIS simulated mean light temperature deviation (y-axis) for all 399 channels (x-axis).
Figure 4e is the CrIS simulated bright temperature standard deviation (y-axis) for all 399 channels (x-axis).
Detailed Description
The technical method of the present invention will be described in detail below with reference to the accompanying drawings, but the scope of the present invention is not limited to the examples.
An infrared brightness temperature simulation method fusing a numerical mode and satellite microwave cloud inversion data comprises the following steps:
preprocessing WRF numerical simulation data and satellite ATMS microwave cloud inversion data;
preprocessing the WRF numerical simulation data, namely preprocessing the simulation data of the WRF numerical weather forecast mode, and extracting conventional atmospheric elements, surface elements and atmospheric cloud water substance elements; preprocessing the satellite ATMS microwave cloud inversion data, namely preprocessing the ATMS microwave cloud inversion data generated by the MiRS system, and extracting atmospheric cloud water substance elements;
the atmospheric conventional elements comprise temperature, water vapor, air pressure and the like; the surface elements comprise surface temperature, surface air pressure and the like; the atmospheric cloud water substance elements comprise a cloud water mixing ratio, a cloud ice mixing ratio, a rainwater mixing ratio and the like;
step two, performing quality control on satellite micro ATMS microwave cloud inversion data; the method for controlling the quality of the satellite micro ATMS microwave cloud inversion data comprises the steps of significant error detection, background field outlier detection and space consistency detection;
1) and (3) major error detection:
cloud-water mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000091
Integrating in vertical direction to obtain the path of cloud water
Figure BDA0003033378120000101
Wherein the subscript c represents the cloud water, and the superscript o represents the observation, which is denoted as the actual integration, CVI for short;
cloud ice mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000102
Integrating in vertical direction to obtain the path of ice cloud
Figure BDA0003033378120000103
Wherein the subscript i represents cloud ice, and the superscript o represents observation, which is denoted as actual integration, abbreviated as IVI;
rainwater mixing ratio of satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000104
Integrating in the vertical direction to obtain the rain path
Figure BDA0003033378120000105
Wherein the subscript r represents Rain water, and the superscript o represents observation, which is marked as Rain vertical integration, RVI for short;
in the three formulas, i represents the track crossing direction, j represents the track direction, k is vertical layering, rho represents the atmospheric density, z represents the vertical height, and kmax represents the vertical serial number corresponding to the maximum height value;
the normal value range of CVI, IVI and RVI is found to be 0-3000g m through sample statistics-2And thus, a critical error checking threshold gamma is setgross＝3000g·m-2；
If CVIo(i,j)＞γgross,or,CVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200001011
Data;
if IVIo(i,j)＞γgross,or,IVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200001012
Data;
if RVIo(i,j)＞γgross,or,RVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure BDA00030333781200001013
Data;
2) background field outlier verification test:
firstly, the cloud-water mixing ratio of WRF numerical mode analog data
Figure BDA0003033378120000106
Cloud ice mixing ratio
Figure BDA0003033378120000107
Rainwater mixing ratio
Figure BDA0003033378120000108
Performing space linear interpolation to obtain satellite ATMS microwave cloud inversion data
Figure BDA0003033378120000109
And
Figure BDA00030333781200001010
wherein subscript c represents cloud water, subscript i represents cloud ice, subscript r represents rain water, and superscript m represents simulation;
secondly, vertical integration is carried out to obtain the cloud water path CVI of the WRF numerical simulation datam(i, j), cloud ice pathway IVIm(i, j) and rainwater Path RVIm(i, j); calculating standard deviation sigma of cloud water path, cloud ice path and rainwater path according to the statistical sampleCVI、σCVIAnd σCVIWherein, in the step (A),
Figure BDA0003033378120000111
Figure BDA0003033378120000112
Figure BDA0003033378120000113
again, an outlier check is performed:
if CVIo(i,j)-CVIm(i,j)|＞3·σCVIThen eliminate the space point (i, j), all height layers
Figure BDA0003033378120000119
Data;
if IVIo(i,j)-IVIm(i,j)|＞3·σIVIThen eliminate the space point (i, j), all height layers
Figure BDA00030333781200001110
Data;
if RVIo(i,j)-RVIm(i,j)|＞3·σRVIThen eliminate the space point (i, j), all height layers
Figure BDA00030333781200001111
Data;
3) and (3) checking the spatial consistency:
first, the CVI of each spatial point (i, j) was calculated separately at a horizontal distance of 110km, approximately 1 degree latitude and longitude, as the radius of the spatial consistency testo(i,j)、IVIo(i, j) and RVIo(i, j) and the average value within a radius of 110km around (i, j)
Figure BDA0003033378120000114
And
Figure BDA0003033378120000115
the differences between, respectively noted:
Figure BDA0003033378120000116
Figure BDA0003033378120000117
Figure BDA0003033378120000118
second, a spatial consistency check is performed:
if deltaCVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200001112
Data;
if deltaIVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200001113
Data;
if deltaRVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure BDA00030333781200001114
And (4) data.
And thirdly, carrying out data fusion on the WRF numerical simulation data and the satellite ATMS microwave cloud inversion data after quality control to obtain a fusion analysis field of the atmospheric cloud water substance elements, including a fusion analysis field of atmospheric cloud water, cloud ice and rainwater.
The fusion algorithm for performing data fusion on the WRF numerical simulation data and the satellite ATMS microwave cloud inversion data after quality control is specifically as follows:
reading WRF numerical mode analog data
Figure BDA0003033378120000121
And
Figure BDA0003033378120000122
satellite ATMS microwave inversion data
Figure BDA0003033378120000123
And
Figure BDA0003033378120000124
and converting into a one-dimensional vector; using variational analysis as a mathematical method of an optimized fusion algorithm, constructing a functional by using cloud water, cloud ice and rainwater data which are subjected to WRF numerical simulation and cloud water, cloud ice and rainwater data which are subjected to satellite ATMS microwave inversion, and solving an extreme value of the functional by using a conjugate gradient descent method so as to solve an optimal fusion analysis field of the cloud water, the cloud ice and the rainwater;
establishing a fusion algorithm of WRF numerical simulation data and satellite ATMS microwave cloud inversion data, wherein the specific algorithm is as follows:
definition of functional Jc、JiAnd Jr，
Figure BDA0003033378120000125
Figure BDA0003033378120000126
Figure BDA0003033378120000127
Wherein three-dimensional variables are combined
Figure BDA0003033378120000128
Is marked as oneDimensional vector form
Figure BDA0003033378120000129
To make three-dimensional variable
Figure BDA00030333781200001210
Is recorded as a one-dimensional vector form
Figure BDA00030333781200001211
To make three-dimensional variable
Figure BDA00030333781200001212
Is recorded as a one-dimensional vector form
Figure BDA00030333781200001213
Wherein the content of the first and second substances,
Figure BDA00030333781200001214
and
Figure BDA00030333781200001215
has a dimension of nx1×ny1×nz1，nx1、ny1And nz1Respectively representing the total number of lattice points of the WRF numerical mode simulation data in the east-west direction, the south-north direction and the vertical direction;
to make three-dimensional variable
Figure BDA00030333781200001216
Is recorded as a one-dimensional vector form
Figure BDA00030333781200001217
To make three-dimensional variable
Figure BDA00030333781200001218
Is recorded as a one-dimensional vector form
Figure BDA00030333781200001219
To make three-dimensional variable
Figure BDA00030333781200001220
Is marked as one dimensionVector form
Figure BDA00030333781200001221
Wherein the content of the first and second substances,
Figure BDA00030333781200001222
and
Figure BDA00030333781200001223
has a dimension of nx2×ny2×nz2，nx2、ny2And nz2Respectively representing the total number of grid points of the satellite ATMS microwave inversion data in the cross-track direction, the along-track direction and the vertical direction;
is prepared from flos Chrysanthemi IndicicError covariance matrix recorded as numerical mode cloud and water mixing ratio data
Figure BDA00030333781200001224
ΜiError covariance matrix recorded as numerical mode cloud ice mixing ratio data
Figure BDA00030333781200001225
ΜrError covariance matrix recorded as numerical mode rainwater mixing ratio data
Figure BDA00030333781200001226
Wherein the content of the first and second substances,
Figure BDA00030333781200001227
and
Figure BDA00030333781200001228
has a dimension of nx1×ny1×nz1；Μc、ΜiAnd μmrHas a dimension of (n)x1×ny1×nz1)2；
Mixing O withcError covariance matrix recorded as satellite microwave cloud and water mixing ratio data
Figure BDA0003033378120000131
OiRecording deviceError covariance matrix of star microwave cloud ice mixing ratio data
Figure BDA0003033378120000132
OrError covariance matrix recorded as satellite microwave rainwater mixing ratio data
Figure BDA0003033378120000133
Wherein the content of the first and second substances,
Figure BDA0003033378120000134
and
Figure BDA0003033378120000135
has a dimension of nx2×ny2×nz2；Oc、OiAnd OrHas a dimension of (n)x2×ny2×nz2)2；
In the formulas (7) - (9), H is an interpolation operator for converting the numerical mode data space into the satellite microwave data space, and superscript T represents matrix transposition;
the core solving process of the fusion algorithm is to Jc、JiAnd JrMinimization of the functional to obtain Qc、QiAnd QrThe fusion analysis field is the optimal fusion analysis field.
A minimum value process is illustrated by taking a functional minimum value solution of the cloud-water mixing ratio as an example;
the functional minimum value solution of the cloud-water mixing ratio is as follows:
the following equivalence relations are established,
ΔQ＝Q-Qm (10)
d＝H(Qm)-Qo (11)
then equation (7) can be rewritten as an incremental form with the subscript c hidden, which is equivalent to,
Figure BDA0003033378120000136
where H is the first order differential form of H, taking into account the first order Taylor's expansion approximation, i.e.
Figure BDA0003033378120000139
Decomposing the error covariance matrix (Md) of the pattern data into a horizontal covariance matrix (D)hAnd vertical covariance matrix DvI.e. by
Figure BDA0003033378120000137
Then the equation (12) is equivalent to,
Figure BDA0003033378120000138
the gradient of the equation (13) to the delta Q is obtained,
▽J(ΔQ)＝(DhDv)-1ΔQ+HTO-1(HΔQ-d) (14)
before solving the gradient descent algorithm, preprocessing is carried out, x is recorded as,
x＝(DhDv)-1ΔQ (15)
then equation (14) is equivalent to:
▽J(x)＝x+HTO-1(HDhDvx-d) (16)
solving gradient descent for ^ J (x) by adopting a classical conjugate gradient descent method, converging a conjugate gradient descent iteration algorithm when a value of ^ J (x) is iterated to 1/1000 of the initial ^ J (x), wherein x corresponding to convergence is a vector corresponding to the minimization of a functional;
finally, Δ Q can be calculated from x by equation (15), and Q can be calculated from Δ Q by equation (10), where Q is a one-dimensional vector corresponding to the final fusion analysis field, and transposing Q into a three-dimensional field is the final result of the fusion algorithm, i.e., the fusion analysis field, because Q is the same as Qc、QiAnd QrThe variational analysis algorithms of (a) are the same and are therefore denoted collectively by the symbol Q.
Extracting atmospheric conventional elements of the WRF numerical simulation data and atmospheric cloud water substances of the fusion analysis field as vertical profiles, and taking the vertical profiles as input profiles of the radiation transmission model CRTM; extracting earth surface elements of the WRF numerical simulation data as single-point elements, inputting the single-point elements and geometric parameters corresponding to the satellite field of view into a radiation transmission model CRTM, and calculating satellite infrared simulation brightness temperature on each grid point; and converting the infrared simulated brightness temperature value of the satellite into a CrIS view field by adopting a CrIS observation and grid data view field matching technology, calculating the weight of each grid simulated value in a CrIS view field range through a two-dimensional Gaussian distribution function, and further obtaining the average simulated brightness temperature through weighted average to serve as the final CrIS view field simulated brightness temperature.
The weight for a corresponding lattice point calculated by a two-dimensional gaussian-shaped function is expressed as follows:
Figure BDA0003033378120000141
wherein x and y represent the horizontal coordinates of the grid field, x0And y0Represents the horizontal coordinate corresponding to the center of CrIS field of view, LxAnd LyThe characteristic scale is represented, and w (x, y) represents the weight corresponding to the (x, y) grid point.
Compared with the traditional infrared bright temperature simulation scheme, the improvement is as follows:
1) the vertical profile of the cloud water substance element input into the radiation transmission model CRTM is derived from an analysis field of a fusion algorithm, and is not WRF numerical simulation data;
2) the geometric characteristics of the CrIS observation field are further considered, a space matching technology of the CrIS observation field and a lattice field is developed, the weight of each lattice simulation value in the CrIS field range is calculated through a two-dimensional Gaussian distribution function, the average simulated bright temperature is obtained through weighted average, and the simulated bright temperature corresponding to the CrIS field coordinate position is not calculated through simple linear interpolation or nearest neighbor interpolation.
FIG. 1a is a traditional infrared brightness temperature simulation method, and FIG. 1b is a flow chart of the infrared brightness temperature simulation method for fusing a numerical model and satellite microwave cloud inversion data according to the invention; the comparison result shows that: (1) according to the traditional infrared bright temperature simulation method, atmospheric conventional elements, surface elements and cloud water substance elements are extracted from numerical mode simulation data, processed into a vertical profile form according to a grid point sequence, input into a radiation transmission model, satellite infrared simulated bright temperature on each grid point is calculated, and then simulated bright temperature corresponding to a CrIS observation visual field is obtained through interpolation. (2) Compared with the traditional infrared bright temperature simulation method, the method has the advantages that firstly, cloud water substance elements simulated in a numerical mode and cloud water substance element data inverted by satellite microwaves are subjected to fusion analysis, the vertical profile of the cloud water substance elements is extracted from a fusion analysis field, then the vertical profile is input into a radiation transmission model, and the infrared simulated bright temperature is calculated; and secondly, performing field matching on the CrIS observation field and the lattice data, calculating the weight of each lattice analog value in the CrIS field range through a two-dimensional Gaussian function, and calculating the final CrIS field analog brightness temperature through weighted average.
Fig. 2 is a comparison graph of field effects of cloud water, cloud ice and rain analysis of numerical model simulation data, satellite microwave inversion data and fusion algorithm analysis data, and fig. 2a-2i are enlarged views of corresponding portions in fig. 2.
Fig. 2 shows cloud water, cloud ice and rainwater distribution characteristics analyzed by three analysis methods of numerical model simulation data, satellite microwave inversion data and fusion algorithm, and shows the effect of fusion analysis: for convenience of analysis, vertical integral quantities of cloud water, cloud ice and rainwater data are compared, and in the case of typhoon "lingering" in 9/6/2019, compared with numerical simulation data, fused data are better in structure depiction of a typhoon eye wall (about 30 degrees N and 125 degrees E) and a south side peripheral rain zone (about 20 degrees N) compared with numerical simulation data, cloud ice and rainwater levels of the south side rain zone are obviously enhanced, the fused data are closer to satellite microwave inversion data, and the effect of satellite microwave actual observation is fully reflected.
FIG. 3a is a schematic diagram of matching of an observation field under a CrIS star with a grid field, a gray circle representing the boundary of the CrIS field, and a black dot representing the position of a grid; FIG. 3b is a Gaussian weight of each bin analog value within the CrIS field of view; in the present invention, the illumination temperature simulation values corresponding to the CrIS field are obtained by performing a weighted average on the illumination temperature simulation values of the grid within the CrIS observation field geometric range of FIG. 3a, wherein the weight contribution of each grid is calculated by a two-dimensional Gaussian function, and the result is shown in FIG. 3 b.
CrIS channels 114 (wave number: 895.625 cm) corresponding to 06 days (UTC) typhoon in 2019, 9, 6 and 6-1) FIG. 4a is a spatial distribution diagram of simulated infrared brightness temperature using a conventional infrared brightness temperature simulation method; FIG. 4b is a spatial distribution diagram of simulated infrared brightness temperature by the method of the present invention, and FIG. 4c is a spatial distribution diagram of CrIS observed brightness temperature. Fig. 4d is the CrIS simulated mean bright temperature deviation (y-axis) for all 399 channels (x-axis), fig. 4e is the CrIS simulated standard bright temperature deviation (y-axis) for all 399 channels (x-axis), where the gray triangles represent the traditional infrared bright temperature simulation method, the black circles represent the method of the present invention, and the sample statistics period ranges from 8 months 25 days in 2019 to 9 months 7 days in 2019.
Comparing fig. 4 a-4 e, the comparison between the simulated infrared bright temperature and the CrIS observed bright temperature, and the simulated average deviation of bright temperature and standard deviation of bright temperature for all 399 CrIS channels are shown: for the typhoon 'lingering' cases of 06 days (UTC) in 9, 6 and 2019, taking CrIS channel 114 (wave number: 895.625cm-1) as an example, the effect of the traditional infrared brightness temperature simulation method on infrared brightness temperature simulation is obviously higher at the periphery of a southern side of the typhoon with a jade belt (20 degrees N), namely, the rain belt range corresponding to the low brightness temperature value is smaller; the method provided by the invention can well improve the simulation effect of the area, and the brightness and temperature are closer to those of CrIS observation. The long-time sample statistical results from 25/8/2019 to 7/9/2019 show that compared with the conventional infrared brightness temperature simulation method, the brightness temperature simulation deviation is remarkably reduced (closer to 0), 399 channels in the CrIS infrared band are remarkably improved, and the standard deviation of brightness temperature simulation is also remarkably reduced, which shows that the infrared brightness temperature simulation method fusing numerical mode simulation data and satellite microwave inversion data has higher simulation precision.
As noted above, while the present invention has been shown and described with reference to certain preferred embodiments, it is not to be construed as limited thereto. Various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (6)

1. An infrared brightness temperature simulation method fusing a numerical mode and satellite microwave cloud inversion data is characterized by comprising the following steps:
preprocessing WRF numerical simulation data and satellite ATMS microwave cloud inversion data;
preprocessing the WRF numerical simulation data, namely preprocessing the simulation data of the WRF numerical weather forecast mode, and extracting conventional atmospheric elements, surface elements and atmospheric cloud water substance elements;
preprocessing the satellite ATMS microwave cloud inversion data, namely preprocessing the ATMS microwave cloud inversion data generated by the MiRS system, and extracting atmospheric cloud water substance elements;
step two, performing quality control on satellite micro ATMS microwave cloud inversion data;
performing data fusion on WRF numerical simulation data and satellite ATMS microwave cloud inversion data after quality control to obtain a fusion analysis field of atmospheric cloud and water substance elements;
extracting atmospheric conventional elements of the WRF numerical simulation data and atmospheric cloud water substances of the fusion analysis field as vertical profiles, and taking the vertical profiles as input profiles of the radiation transmission model CRTM; extracting earth surface elements of the WRF numerical simulation data as single-point elements, inputting the single-point elements and geometric parameters corresponding to the satellite field of view into a radiation transmission model CRTM, and calculating satellite infrared simulation brightness temperature on each grid point; and converting the infrared simulated brightness temperature value of the satellite into a CrIS view field by adopting a CrIS observation and grid data view field matching technology, calculating the weight of each grid simulated value in a CrIS view field range through a two-dimensional Gaussian distribution function, and further obtaining the average simulated brightness temperature through weighted average to serve as the final CrIS view field simulated brightness temperature.
2. The method for simulating the infrared brightness and temperature of the fused numerical mode and satellite microwave cloud inversion data according to claim 1, wherein the atmospheric conventional elements comprise temperature, water vapor and air pressure; the surface elements comprise surface temperature and surface air pressure; the atmospheric cloud water substance elements comprise a cloud water mixing ratio, a cloud ice mixing ratio and a rainwater mixing ratio.
3. The infrared brightness temperature simulation method for fusing the numerical mode and the satellite microwave cloud inversion data according to claim 2, wherein the method for performing quality control on the satellite micro ATMS microwave cloud inversion data in the second step comprises a major error test, a background field outlier test and a space consistency test;
1) and (3) major error detection:
cloud-water mixing ratio of satellite ATMS microwave cloud inversion data
Figure FDA0003033378110000011
Integrating in vertical direction to obtain the path of cloud water
Figure FDA0003033378110000012
Wherein the subscript c represents the cloud water, and the superscript o represents the observation, which is denoted as the actual integration, CVI for short;
cloud ice mixing ratio of satellite ATMS microwave cloud inversion data
Figure FDA0003033378110000013
Integrating in vertical direction to obtain the path of ice cloud
Figure FDA0003033378110000021
Wherein the subscript i represents cloud ice, and the superscript o represents observation, which is denoted as actual integration, abbreviated as IVI;
rainwater mixing ratio of satellite ATMS microwave cloud inversion data
Figure FDA0003033378110000022
Integrating in the vertical direction to obtain the rain path
Figure FDA0003033378110000023
Wherein the subscript r represents rain, the superscript o represents observation and recordingIs Rain vertical integration, abbreviated as RVI;
in the three formulas, i represents the track crossing direction, j represents the track direction, k is vertical layering, rho represents the atmospheric density, z represents the vertical height, and kmax represents the vertical serial number corresponding to the maximum height value;
the normal value range of CVI, IVI and RVI is found to be 0-3000g m through sample statistics-2And thus, a critical error checking threshold gamma is setgross＝3000g·m-2；
If CVIo(i,j)＞γgross,or,CVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure FDA00030333781100000212
Data;
if IVIo(i,j)＞γgross,or,IVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure FDA00030333781100000213
Data;
if RVIo(i,j)＞γgross,or,RVIoIf (i, j) < 0, then eliminating the space point (i, j), all height layers
Figure FDA00030333781100000214
Data;
2) background field outlier verification test:
firstly, the cloud-water mixing ratio of WRF numerical mode analog data
Figure FDA0003033378110000024
Cloud ice mixing ratio
Figure FDA0003033378110000025
Rainwater mixing ratio
Figure FDA0003033378110000026
The linear interpolation of the space is carried out,interpolation to satellite ATMS microwave cloud inversion data
Figure FDA0003033378110000027
Figure FDA0003033378110000028
And
Figure FDA0003033378110000029
wherein subscript c represents cloud water, subscript i represents cloud ice, subscript r represents rain water, and superscript m represents simulation;
secondly, vertical integration is carried out to obtain the cloud water path CVI of the WRF numerical simulation datam(i, j), cloud ice pathway IVIm(i, j) and rainwater Path RVIm(i, j); calculating standard deviation sigma of cloud water path, cloud ice path and rainwater path according to the statistical sampleCVI、σCVIAnd σCVIWherein, in the step (A),
Figure FDA00030333781100000210
Figure FDA00030333781100000211
Figure FDA0003033378110000031
again, an outlier check is performed:
if CVIo(i,j)-CVIm(i,j)|＞3·σCVIThen eliminate the space point (i, j), all height layers
Figure FDA0003033378110000032
Data;
if IVIo(i,j)-IVIm(i,j)|＞3·σIVIThen pick outAll height levels except the spatial point (i, j)
Figure FDA0003033378110000033
Data;
if RVIo(i,j)-RVIm(i,j)|＞3·σRVIThen eliminate the space point (i, j), all height layers
Figure FDA0003033378110000034
Data;
3) and (3) checking the spatial consistency:
first, the CVI of each spatial point (i, j) is calculated separately with the horizontal distance of 110km as the radius of the spatial consistency checko(i,j)、IVIo(i, j) and RVIo(i, j) and the average value within a radius of 110km around (i, j)
Figure FDA0003033378110000035
And
Figure FDA0003033378110000036
the differences between, respectively noted:
Figure FDA0003033378110000037
Figure FDA0003033378110000038
Figure FDA0003033378110000039
second, a spatial consistency check is performed:
if deltaCVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure FDA00030333781100000310
Data;
if deltaIVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure FDA00030333781100000311
Data;
if deltaRVI(i,j)＞100g·m-2Then eliminate the space point (i, j), all height layers
Figure FDA00030333781100000312
And (4) data.
4. The method for simulating the infrared brightness temperature of the fused numerical mode and satellite microwave cloud inversion data according to claim 3, wherein a fusion algorithm for performing data fusion on the WRF numerical simulation data and the satellite ATMS microwave cloud inversion data after quality control in the third step is specifically as follows:
reading WRF numerical mode analog data
Figure FDA00030333781100000313
And
Figure FDA00030333781100000314
satellite ATMS microwave inversion data
Figure FDA00030333781100000315
And
Figure FDA00030333781100000316
and converting into a one-dimensional vector; using variational analysis as a mathematical method of an optimized fusion algorithm, constructing a functional by using cloud water, cloud ice and rainwater data which are subjected to WRF numerical simulation and cloud water, cloud ice and rainwater data which are subjected to satellite ATMS microwave inversion, and solving an extreme value of the functional by using a conjugate gradient descent method so as to solve an optimal fusion analysis field of the cloud water, the cloud ice and the rainwater;
the specific algorithm is as follows:
definition of functional Jc、JiAnd Jr，
Figure FDA0003033378110000041
Figure FDA0003033378110000042
Figure FDA0003033378110000043
Wherein three-dimensional variables are combined
Figure FDA0003033378110000044
Is recorded as a one-dimensional vector form
Figure FDA0003033378110000045
To make three-dimensional variable
Figure FDA0003033378110000046
Is recorded as a one-dimensional vector form
Figure FDA0003033378110000047
To make three-dimensional variable
Figure FDA0003033378110000048
Is recorded as a one-dimensional vector form
Figure FDA0003033378110000049
Wherein the content of the first and second substances,
Figure FDA00030333781100000410
and
Figure FDA00030333781100000411
has a dimension of nx1×ny1×nz1，nx1、ny1And nz1Respectively representing the total number of lattice points of the WRF numerical mode simulation data in the east-west direction, the south-north direction and the vertical direction;
to make three-dimensional variable
Figure FDA00030333781100000412
Is recorded as a one-dimensional vector form
Figure FDA00030333781100000413
To make three-dimensional variable
Figure FDA00030333781100000414
Is recorded as a one-dimensional vector form
Figure FDA00030333781100000415
To make three-dimensional variable
Figure FDA00030333781100000416
Is recorded as a one-dimensional vector form
Figure FDA00030333781100000417
Wherein the content of the first and second substances,
Figure FDA00030333781100000418
and
Figure FDA00030333781100000419
has a dimension of nx2×ny2×nz2，nx2、ny2And nz2Respectively representing the total number of grid points of the satellite ATMS microwave inversion data in the cross-track direction, the along-track direction and the vertical direction;
is prepared from flos Chrysanthemi IndicicError covariance matrix recorded as numerical mode cloud and water mixing ratio data
Figure FDA00030333781100000420
ΜiIs recorded as a numerical patternError covariance matrix of cloud ice mixing ratio data
Figure FDA00030333781100000421
ΜrError covariance matrix recorded as numerical mode rainwater mixing ratio data
Figure FDA00030333781100000422
Wherein the content of the first and second substances,
Figure FDA00030333781100000423
and
Figure FDA00030333781100000424
has a dimension of nx1×ny1×nz1；Μc、ΜiAnd μmrHas a dimension of (n)x1×ny1×nz1)2；
Mixing O withcError covariance matrix recorded as satellite microwave cloud and water mixing ratio data
Figure FDA00030333781100000425
OiError covariance matrix recorded as satellite microwave cloud ice mixing ratio data
Figure FDA00030333781100000426
OrError covariance matrix recorded as satellite microwave rainwater mixing ratio data
Figure FDA00030333781100000427
Wherein the content of the first and second substances,
Figure FDA00030333781100000428
and
Figure FDA00030333781100000429
has a dimension of nx2×ny2×nz2；Oc、OiAnd OrHas a dimension of(nx2×ny2×nz2)2；
In the formulas (7) - (9), H is an interpolation operator for converting the numerical mode data space into the satellite microwave data space, and superscript T represents matrix transposition;
the core solving process of the fusion algorithm is to Jc、JiAnd JrMinimization of the functional to obtain Qc、QiAnd QrThe fusion analysis field is the optimal fusion analysis field.
5. The method for simulating the infrared brightness and temperature of the fused numerical mode and satellite microwave cloud inversion data according to claim 4, wherein the solution of the functional minimum of the cloud-water mixing ratio is as follows:
the following equivalence relations are established,
ΔQ＝Q-Qm (10)
d＝H(Qm)-Qo (11)
then equation (7) can be rewritten as an incremental form with the subscript c hidden, which is equivalent to,
Figure FDA0003033378110000051
where H is the first order differential form of H, taking into account the first order Taylor's expansion approximation, i.e.
Figure FDA0003033378110000059
Decomposing the error covariance matrix (Md) of the pattern data into a horizontal covariance matrix (D)hAnd vertical covariance matrix DvI.e. by
Figure FDA0003033378110000052
Then the equation (12) is equivalent to,
Figure FDA0003033378110000053
the gradient of the equation (13) to the delta Q is obtained,
Figure FDA0003033378110000054
before solving the gradient descent algorithm, preprocessing is carried out, x is recorded as,
x＝(DhDv)-1ΔQ (15)
then equation (14) is equivalent to:
Figure FDA0003033378110000055
using a classical conjugate gradient descent method
Figure FDA0003033378110000056
The gradient is decreased when
Figure FDA0003033378110000057
Iterating the value to the initial
Figure FDA0003033378110000058
1/1000, converging the conjugate gradient descent iterative algorithm, wherein x corresponding to convergence is a vector corresponding to the minimization of the functional;
finally, Δ Q can be calculated from x by equation (15), and Q can be calculated from Δ Q by equation (10), where Q is a one-dimensional vector corresponding to the final fusion analysis field, and transposing Q into a three-dimensional field is the final result of the fusion algorithm, i.e., the fusion analysis field, because Q is the same as Qc、QiAnd QrThe variational analysis algorithms of (a) are the same and are therefore denoted collectively by the symbol Q.
6. The method for simulating the infrared brightness temperature of the fused numerical mode and satellite microwave cloud inversion data according to claim 1, wherein the weight expression of the corresponding grid points calculated by the two-dimensional Gaussian function in the fourth step is as follows:
Figure FDA0003033378110000061
wherein x and y represent the horizontal coordinates of the grid field, x0And y0Represents the horizontal coordinate corresponding to the center of CrIS field of view, LxAnd LyThe characteristic scale is represented, and w (x, y) represents the weight corresponding to the (x, y) grid point.
CN202110436628.6A 2021-04-22 2021-04-22 Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data Active CN113111529B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110436628.6A CN113111529B (en) 2021-04-22 2021-04-22 Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110436628.6A CN113111529B (en) 2021-04-22 2021-04-22 Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data

Publications (2)

Publication Number Publication Date
CN113111529A true CN113111529A (en) 2021-07-13
CN113111529B CN113111529B (en) 2023-10-10

Family

ID=76719430

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110436628.6A Active CN113111529B (en) 2021-04-22 2021-04-22 Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data

Country Status (1)

Country Link
CN (1) CN113111529B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113591387A (en) * 2021-08-05 2021-11-02 安徽省气象台 Huber norm constraint-based satellite data inversion precipitation method and system
CN116341352A (en) * 2022-07-27 2023-06-27 南京气象科技创新研究院 Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103675792A (en) * 2013-12-06 2014-03-26 中国科学院遥感与数字地球研究所 Method for quickly inversing ozone profile of CrIS hyper-spectral data
CN108614309A (en) * 2018-05-11 2018-10-02 西安理工大学 Cloud liquid water path detection system and method
CN108875254A (en) * 2018-07-03 2018-11-23 南京信息工程大学 A kind of One-Dimensional Variational inversion method of Atmosphere and humidity profiles
US20190331831A1 (en) * 2018-04-26 2019-10-31 Jiangsu Meteorological Observatory Sea fog monitoring method based on multi-source satellite remote sensing data
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
CN112197865A (en) * 2020-09-02 2021-01-08 华中科技大学 Estimation method and system for observation brightness temperature data error of satellite-borne microwave radiometer
CN112213727A (en) * 2020-10-15 2021-01-12 国家卫星气象中心(国家空间天气监测预警中心) Precipitation correction method of satellite-borne radar based on active and passive microwave combined detection
CN112327388A (en) * 2020-02-13 2021-02-05 中国农业科学院农业资源与农业区划研究所 All-weather relative humidity estimation method based on full remote sensing data

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103675792A (en) * 2013-12-06 2014-03-26 中国科学院遥感与数字地球研究所 Method for quickly inversing ozone profile of CrIS hyper-spectral data
US20190331831A1 (en) * 2018-04-26 2019-10-31 Jiangsu Meteorological Observatory Sea fog monitoring method based on multi-source satellite remote sensing data
CN108614309A (en) * 2018-05-11 2018-10-02 西安理工大学 Cloud liquid water path detection system and method
US20200064514A1 (en) * 2018-05-11 2020-02-27 Xi'an University Of Technology Cloud water resource detecting system and method
CN108875254A (en) * 2018-07-03 2018-11-23 南京信息工程大学 A kind of One-Dimensional Variational inversion method of Atmosphere and humidity profiles
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
CN112327388A (en) * 2020-02-13 2021-02-05 中国农业科学院农业资源与农业区划研究所 All-weather relative humidity estimation method based on full remote sensing data
CN112197865A (en) * 2020-09-02 2021-01-08 华中科技大学 Estimation method and system for observation brightness temperature data error of satellite-borne microwave radiometer
CN112213727A (en) * 2020-10-15 2021-01-12 国家卫星气象中心(国家空间天气监测预警中心) Precipitation correction method of satellite-borne radar based on active and passive microwave combined detection

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
VENKATA B. DODLA: "Prediction of tropical cyclone over North Indian Ocean using WRF model: sensitivity to scatterometer winds, ATOVS and ATMS radiances", 《PROCEEDINGS VOLUME 9882, REMOTE SENSING AND MODELING OF THE ATMOSPHERE, OCEANS, AND INTERACTIONS VI》 *
刘乐; 李华实: "利用RTTOV模拟台风"海燕"卫星红外通道亮温的研究", 《第33届中国气象学会年会 S1 灾害天气监测、分析与预报 》, pages 941 - 951 *
陈震霆: "基于卫星偏振辐射信息的云相态空协同反演研究", 《中国博士学位论文全文数据库 基础科学辑》, pages 009 - 3 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113591387A (en) * 2021-08-05 2021-11-02 安徽省气象台 Huber norm constraint-based satellite data inversion precipitation method and system
CN113591387B (en) * 2021-08-05 2023-09-01 安徽省气象台 Satellite data inversion precipitation method and system based on Huber norm constraint
CN116341352A (en) * 2022-07-27 2023-06-27 南京气象科技创新研究院 Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint
CN116341352B (en) * 2022-07-27 2023-10-24 南京气象科技创新研究院 Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint

Also Published As

Publication number Publication date
CN113111529B (en) 2023-10-10

Similar Documents

Publication Publication Date Title
CN110174359B (en) Aviation hyperspectral image soil heavy metal concentration assessment method based on Gaussian process regression
CN109344865B (en) Data fusion method for multiple data sources
CN112213727B (en) Precipitation correction method of satellite-borne radar based on active and passive microwave combined detection
Oshchepkov et al. Effects of atmospheric light scattering on spectroscopic observations of greenhouse gases from space. Part 2: Algorithm intercomparison in the GOSAT data processing for CO2 retrievals over TCCON sites
CN113111529B (en) Infrared bright temperature simulation method integrating numerical mode and satellite microwave cloud inversion data
CN112699959B (en) Multi-source multi-scale precipitation data fusion method and device based on energy functional model
CN111579504A (en) Atmospheric pollution component vertical distribution inversion method based on optical remote sensing
Tian et al. Improving surface soil moisture retrievals through a novel assimilation algorithm to estimate both model and observation errors
Marseille et al. Toward scatterometer winds assimilation in the mesoscale HARMONIE model
Zhang et al. Assimilation of MWHS and MWTS radiance data from the FY-3A satellite with the POD-3DEnVar method for forecasting heavy rainfall
CN114880933A (en) Atmospheric temperature and humidity profile inversion method and system for non-exploration-site foundation microwave radiometer based on reanalysis data
Slingo et al. Simulation of clear‐sky outgoing longwave radiation over the oceans using operational analyses
Chen et al. Data assimilation of adaptive observation and application for typhoon forecasts over the Western North Pacific
Shen et al. Assimilation of Himawari-8 imager radiance data with the WRF-3DVAR system for the prediction of Typhoon Soudelor
Lee et al. All‐sky microwave humidity sounder assimilation in the Korean Integrated Model forecast system
CN116822141A (en) Method for inverting optical thickness of night atmospheric aerosol by utilizing satellite micro-optic remote sensing
He et al. Modis aerosol optical thickness product algorithm verification and analysis
CN114047563A (en) All-weather assimilation method of infrared hyperspectrum
Zhang et al. Detailed evolution characteristics of an inclined structure hailstorm observed by polarimetric radar over the South China coast
Zhang et al. Preliminary evaluation of all-sky radiance assimilation scheme with POD-3DEnVar method
Qin et al. Assimilation of FY-3D MWTS-II radiance with 3D precipitation detection and the impacts on typhoon forecasts
CN116341352B (en) Static satellite land infrared bright temperature simulation method based on earth surface temperature observation information constraint
Cai et al. Latitudinal and Scan-dependent Biases of Microwave Humidity Sounder Measurements and Their Dependences on Cloud Ice Water Path
An et al. Ozone Profile Retrieval Algorithm Based on GEOS-Chem Model in the Middle and Upper Atmosphere
Aires Atmospheric water vapour profiling over ocean/land and for clear/cloudy situations using microwave observations

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