Cross-calibration method and system for spaceborne microwave radiometer channel level

Abstract

The application provides a cross-channel calibration method and system for a spaceborne microwave radiometer, and belongs to the fields of microwave remote sensing and on-orbit calibration of spaceborne microwave radiometers. A plurality of natural calibration scenes with complementary brightness temperature characteristics and separated geographical positions are used to jointly calibrate a target load, a reference load and radiation transmission simulation information in a wide brightness temperature range, and independent correction relationships are established for different frequency-polarization channels, so that the coverage of calibration samples is expanded, and the consistency, accuracy and applicability of the corrected brightness temperature observed by different channels are improved. The calibration system comprises the following modules: a natural calibration area determination module; a channel matching module; a data acquisition and processing module; a simulated brightness temperature calculation module; a double-difference bias calculation module; a correction model establishment module; and a brightness temperature correction module. The application can significantly expand the coverage of calibration data, and effectively solve the problems of insufficient brightness temperature range and extrapolation error caused by a single natural scene.

Description

Technical Field

[0001] This application relates to the field of microwave remote sensing and on-orbit calibration of spaceborne microwave radiometers, specifically proposing a channel-level cross-calibration method and system for spaceborne microwave radiometers based on multiple natural scenes and joint double difference. Background Technology

[0002] Currently, spaceborne microwave radiometers are widely used for remote sensing monitoring of geophysical parameters such as sea surface temperature, sea surface wind speed, atmospheric water vapor, cloud water content, soil moisture, and land surface temperature.

[0003] Existing external calibration methods for spaceborne microwave radiometers mainly include direct cross-calibration based on observation matching, simulation calibration based on radiative transfer models, and model-assisted double-difference cross-calibration. Direct cross-calibration typically utilizes spatiotemporally matched observation data between the target and reference payloads, using the observed brightness temperature of the reference payload as a benchmark to correct the target payload. However, due to differences in center frequency, observation angle, spatial resolution, and instrument response characteristics among different payloads, directly comparing observed brightness temperatures can easily introduce additional errors. Calibration methods based on radiative transfer models calculate theoretically simulated brightness temperatures by inputting atmospheric and surface state parameters, and use the difference between the simulated and actual observed brightness temperatures for correction. However, their results are easily affected by errors in the radiative transfer model and uncertainties in the input environmental parameters.

[0004] Existing calibration methods suffer from the following shortcomings: First, they typically rely on single or limited types of natural calibration scenarios, and different natural scenarios can only provide a limited range of brightness temperatures. For example, ocean areas usually provide lower brightness temperature samples, while vegetated areas can provide higher brightness temperature samples. A single natural scenario is insufficient to cover the wide dynamic range of brightness temperature observed by spaceborne microwave radiometers in actual operation. When using calibration relationships established with a limited brightness temperature range to correct data from other brightness temperature regions, extrapolation errors are easily introduced, reducing the applicability of the calibration results. Second, different natural scenarios have different underlying microwave radiation characteristics. The microwave emission mechanisms of oceans, vegetated areas, bare soil or desert areas, and snow-covered areas differ significantly. Using a single natural scenario or a uniform surface radiation model makes it difficult to fully utilize the complementary information between different natural scenarios, limiting the effective use of calibration samples. Moreover, spaceborne microwave radiometers typically contain multiple observation channels with different center frequencies and polarization modes, and the response characteristics of different channels to surface and atmospheric parameters vary. Existing methods typically establish a unified correction relationship at the overall load level, making it difficult to describe the deviation variation characteristics of different frequency-polarization channels separately, which makes it difficult to achieve independent and precise correction for different channels.

[0005] In view of the above, this application is hereby submitted. Summary of the Invention

[0006] The channel-level cross-calibration method and system for spaceborne microwave radiometers proposed in this application aims to solve the problems existing in the prior art by utilizing multiple geographically separated and complementary natural calibration scenarios. Within a wide brightness temperature range, it combines target payload, reference payload, and radiative transfer simulation information to establish independent correction relationships for different frequency-polarization channels. This aims to expand the coverage of calibration samples and effectively reduce calibration deviations caused by limited scenarios and limited brightness temperature ranges, thereby improving the consistency, accuracy, and applicability of brightness temperature correction for different channels.

[0007] To achieve the above design objectives, the spaceborne microwave radiometer channel-level cross-calibration system proposed in this application consists of the following modules: The natural calibration region determination module is used to select multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics to construct a joint calibration sample set. The channel matching module is used to establish the correspondence between the target load calibration channel and the reference load reference channel; The data acquisition and processing module is used to acquire the observed brightness temperature data and environmental parameters of the target load and reference load, and to perform quality control and spatiotemporal matching processing. The simulated brightness temperature calculation module is used to calculate the simulated brightness temperature of the corresponding channels of the target load and the reference load according to the type of the underlying surface in different natural calibration regions. The double-difference deviation calculation module is used to obtain the double-difference deviation amount based on the difference between the simulation-observation deviation of the target load and the simulation-observation deviation of the reference load. The calibration model establishment module is used to combine double-difference bias samples from multiple natural calibration regions to establish a linear calibration relationship between the double-difference bias of each channel to be calibrated and the observed brightness temperature. The brightness temperature correction module is used to calculate the correction amount according to the linear correction relationship and correct the original observed brightness temperature of the target load to obtain the corrected brightness temperature after independent calibration of each channel.

[0008] Based on the aforementioned channel-level cross-calibration system for spaceborne microwave radiometers, this application also proposes a channel-level cross-calibration method for spaceborne microwave radiometers. This method involves selecting multiple geographically separated natural calibration regions with different underlying surface types and complementary brightness temperature characteristics to construct joint calibration samples covering different brightness temperature ranges. For the target payload channel to be calibrated, a corresponding reference channel is determined from the reference payload. The observed brightness temperature and environmental parameters of the corresponding channels for both the target and reference payloads are obtained. Combined with surface microwave emissivity models and atmospheric radiative transfer models for different natural scenarios, the simulated brightness temperature of the corresponding channels for both the target and reference payloads is calculated. A double-difference deviation is constructed based on the deviation between the simulated and actual observed brightness temperatures. A channel-level correction relationship between the double-difference deviation and the observed brightness temperature of the target payload is established using multiple natural scenario samples. This yields independent correction parameters for each frequency-polarization channel, and the observed brightness temperature of the target payload is then corrected.

[0009] Different frequency-polarization channels of the target load have independent corresponding linear correction slopes and linear correction intercepts, and different channels do not share the same correction parameters.

[0010] The method includes the following implementation process: S1. Selection of natural calibration scenarios and construction of joint calibration samples; Multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics are selected. The temporal stability and spatial uniformity of each natural calibration region are screened to ensure that each natural calibration region provides effective calibration samples for different brightness temperature ranges within the target load observation brightness temperature range. The effective calibration samples obtained from multiple natural calibration regions are combined to construct a joint calibration sample set covering the preset brightness temperature dynamic range of the target load. S2. Matching the target channel with the reference channel: For any uncalibrated channel of the target load, a corresponding reference channel is determined from the reference load based on the center frequency, polarization mode, and observation parameters of the uncalibrated channel; a correspondence is established between the uncalibrated channel of the target load and the reference channel of the reference load. S3. Acquisition and preprocessing of observation data and environmental parameters: Acquire the observed brightness temperature data of the target payload and reference payload in the corresponding channels of each natural calibration region, and acquire the time information, spatial location information and environmental parameters corresponding to the observed brightness temperature data; perform quality control, spatiotemporal matching and data processing on the acquired data to obtain effective calibration samples; S4. Model-assisted simulation of brightness temperature calculation; Based on the underlying surface type corresponding to different natural calibration areas, the corresponding simulated brightness temperature calculation model is selected, and combined with the atmospheric radiation transfer model, the simulated brightness temperature corresponding to the target load calibration channel and the reference load reference channel is calculated respectively; the target load and the reference load are simulated according to their respective actual center frequency, polarization mode and observation conditions, so that the obtained simulated brightness temperatures correspond to the actual observation states of the target load and the reference load respectively. S5. Calculation of double difference deviation: For any channel to be calibrated of the target load, calculate the first deviation between the simulated brightness temperature and the actual observed brightness temperature of the target load channel to be calibrated, and the second deviation between the simulated brightness temperature and the actual observed brightness temperature of the reference channel of the reference load; based on the difference between the first deviation and the second deviation, obtain the double difference deviation of the corresponding channel to be calibrated. Calculate the first deviation between the simulated brightness temperature and the actual observed brightness temperature of the target load: ; in, This represents the simulated brightness temperature of the target load channel to be calibrated. This indicates the actual observed brightness temperature of the target payload's uncalibrated channel; Calculate the second deviation between the simulated brightness temperature and the observed brightness temperature of the reference channel for the reference load: ; in, This indicates the simulated brightness temperature of the reference channel of the reference load. This indicates the actual observed brightness temperature of the reference channel of the reference load.

[0011] Construct the double-difference deviation DD between the target load and the reference load: ; Through the above processing, the partial radiative transfer model error and environmental parameter error jointly experienced by the target load and the reference load are reduced in the difference process, thereby obtaining the double difference deviation sample corresponding to the current channel to be calibrated. S6. Establishment of a joint correction model across natural scenes: For the same channel to be calibrated, the double-difference deviation samples obtained from different natural calibration regions are combined; linear regression is performed with the observed brightness temperature of the target load as the independent variable and the corresponding double-difference deviation amount as the dependent variable to establish the linear relationship between the double-difference deviation amount and the observed brightness temperature of the target load, and the correction parameters of the target load channel to be calibrated are obtained. S7, Channel-level brightness temperature correction: Calculate the correction amount corresponding to the target load's uncalibrated channel based on the aforementioned correction mapping relationship; The correction amount is applied to the original observed brightness temperature of the target payload to obtain the corrected brightness temperature after cross-calibration. Steps S2 to S7 are executed for different frequencies and polarization channels of the target payload to obtain the independent calibration parameters for each channel to be calibrated, thereby achieving channel-level cross-calibration of the spaceborne microwave radiometer.

[0012] The S1 mentioned above refers to a natural calibration region that includes two or more natural scenes with different microwave radiation characteristics, temporal stability and spatial uniformity. The brightness temperature coverage of the channel to be calibrated is expanded by the brightness temperature complementarity between different natural scenes.

[0013] The matching criteria for the target load calibration channel and the reference load reference channel in S2 include: the center frequency being the same or approximately the same, the polarization mode being consistent or corresponding, and the observation angle being the same or corrected by radiation transfer simulation based on the difference in observation angle.

[0014] The S3 mentioned above includes environmental parameters that describe atmospheric conditions and parameters that describe surface conditions; The atmospheric state parameters mentioned above include one or more of temperature, humidity, and cloud liquid water content; the surface state parameters include one or more of surface temperature, soil moisture content, vegetation parameters, sea surface state parameters, and ice and snow parameters.

[0015] The S4 mentioned above, the atmospheric radiation transfer model, is used to calculate the theoretical simulated brightness temperature of the corresponding channel based on environmental parameters, observation frequency, polarization mode, and observation angle; the simulated brightness temperature calculation model includes the atmospheric radiation transfer model and the surface microwave emissivity model corresponding to different underlying surface types.

[0016] Step S6 involves using the original observed brightness temperature of the target payload. Using as the independent variable and the corresponding double difference deviation DD as the dependent variable, a linear regression method is used to establish the relationship between the two: ; Where a is the linear correction slope corresponding to the current channel to be calibrated, and b is the linear correction intercept corresponding to the current channel to be calibrated.

[0017] S7, as described above, specifies the brightness temperature correction amount corresponding to the current channel to be calibrated: ; Adding the correction amount to the original observed brightness temperature of the target payload, we obtain the corrected brightness temperature after cross-calibration: ; Right now: ; in, This indicates the corrected brightness temperature of the target load channel after joint double-difference cross-calibration of multiple natural scenes.

[0018] In summary, the beneficial effects and advantages of this application compared with the prior art are as follows: 1. This application uses multiple geographically separated natural calibration regions with complementary brightness temperature characteristics. By combining natural scene samples with different brightness temperature levels and different underlying surface types, the coverage of calibration data is significantly expanded, which can effectively solve the problems of insufficient brightness temperature range and extrapolation error caused by a single natural scene.

[0019] 2. This application can use the corresponding surface microwave emissivity model for different natural scenarios, and combine it with the atmospheric radiative transfer model to calculate the simulated brightness temperature of the target load and reference load respectively, so that the simulation process is more consistent with the actual radiation state under different loads, frequencies and observation conditions.

[0020] 3. This application makes full use of the simulation-observation bias of the target load and the reference load to construct a double-difference bias, which effectively reduces the common model error and environmental parameter uncertainty during the difference process, so as to significantly improve the reliability of the cross-calibration results.

[0021] 4. This application can establish independent correction relationships for different frequency-polarization channels, and significantly improve the precision of calibration results for different channels without using uniform correction parameters to describe the response characteristics of all channels.

[0022] 5. This application achieves channel-level cross-calibration of brightness temperature observed by a spaceborne microwave radiometer by forming a set of correction parameters corresponding to each channel to be calibrated, thereby significantly improving the consistency, accuracy and applicability of the corrected brightness temperature data under different natural scenes and different brightness temperature ranges. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. Some specific embodiments of this application will be described in detail below with reference to the accompanying drawings in an exemplary and non-limiting manner. The same reference numerals in the drawings designate the same or similar parts or components. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale.

[0024] Figure 1 This is a schematic flowchart of the channel-level cross-calibration method for the spaceborne microwave radiometer described in this application; Figure 2The following are schematic diagrams of multiple natural calibration regions and their brightness temperature coverage characteristics in the embodiments; wherein, (a) is a schematic diagram of the characteristics of the equatorial Pacific Ocean, (b) is a schematic diagram of the characteristics of the Antarctic ice sheet region, (c) is a schematic diagram of the characteristics of the Taklamakan Desert region, and (d) is a schematic diagram of the characteristics of the Amazon rainforest region. Figure 3 A schematic diagram illustrating the matching of target load and reference load channels, calculation of simulation-observation bias, and construction of double-difference bias. Figure 4 The above is a schematic diagram comparing the simulated brightness temperature and the actual observed brightness temperature of a representative uncalibrated channel (37GHz-H) in different natural calibration regions in the embodiments; where (a) is a comparison diagram of the Amazon rainforest region, (b) is a comparison diagram of the Taklamakan Desert region, (c) is a comparison diagram of the Antarctic ice sheet region, and (d) is a comparison diagram of the equatorial Pacific Ocean. Figure 5 This is a schematic diagram of fitting the linear relationship between the double-difference deviation and the original observed brightness temperature of the target payload based on joint samples from multiple natural calibration regions for a representative uncalibrated channel (37GHz-H) in the embodiment. Figure 6 This is a schematic diagram comparing the average brightness temperature error of the target load channel before and after cross-calibration with the reference load in the selected Amazon rainforest region in the embodiment. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] Example 1, as Figures 1 to 6 As shown, this application proposes a novel channel-level cross-calibration system for a spaceborne microwave radiometer, which comprises the following modules: The natural calibration region determination module is used to select multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics to construct a joint calibration sample set. The channel matching module is used to establish the correspondence between the target load calibration channel and the reference load reference channel; The data acquisition and processing module is used to acquire the observed brightness temperature data and environmental parameters of the target load and reference load, and to perform quality control and spatiotemporal matching processing. The simulated brightness temperature calculation module is used to calculate the simulated brightness temperature of the corresponding channels of the target load and the reference load according to the type of the underlying surface in different natural calibration regions. The double-difference deviation calculation module is used to obtain the double-difference deviation amount based on the difference between the simulation-observation deviation of the target load and the simulation-observation deviation of the reference load. The calibration model establishment module is used to combine double-difference bias samples from multiple natural calibration regions to establish a linear calibration relationship between the double-difference bias of each channel to be calibrated and the observed brightness temperature. The brightness temperature correction module is used to calculate the correction amount according to the linear correction relationship and correct the original observed brightness temperature of the target load to obtain the corrected brightness temperature after independent calibration of each channel.

[0027] Applying the aforementioned channel-level cross-calibration system for spaceborne microwave radiometers, this embodiment proposes a channel-level cross-calibration method for spaceborne microwave radiometers. This method selects multiple geographically separated natural calibration regions with different underlying surface types and complementary brightness temperature characteristics to construct joint calibration samples covering different brightness temperature ranges. For the target payload channel to be calibrated, the corresponding reference channel is determined from the reference payload. The observed brightness temperature and environmental parameters of the corresponding channels for the target and reference payloads are obtained. Combined with surface microwave emissivity models and atmospheric radiative transfer models for different natural scenarios, the simulated brightness temperature of the corresponding channels for the target and reference payloads is calculated respectively. Based on the deviation between the simulated and actual observed brightness temperatures, a double-difference deviation is constructed. A channel-level correction relationship between the double-difference deviation and the observed brightness temperature of the target payload is established using multiple natural scenario samples. This yields independent correction parameters for each frequency-polarization channel, and the observed brightness temperature of the target payload is then corrected.

[0028] Furthermore, the target load has independent linear correction slopes and linear correction intercepts for different frequency-polarization channels, and different channels do not share the same correction parameters.

[0029] Based on the above method, this embodiment selects the following multi-natural scenario joint double-difference execution channel-level cross-calibration process: Using the HY-2B satellite scanning microwave radiometer (SMR) as the target payload and the GCOM-W1 satellite advanced microwave scanning radiometer (AMSR2) as the reference payload, independent cross-calibration of different frequency-polarization channels of the SMR was performed by combining four types of natural calibration regions: the Amazon rainforest, the equatorial Pacific Ocean, the Taklamakan Desert, and the Antarctic ice sheet. The steps included: S1. Selection of natural calibration scenarios and construction of joint calibration samples; Multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics were selected. The temporal stability and spatial uniformity of each natural calibration region were screened so that different natural calibration regions could provide effective calibration samples for different brightness temperature ranges within the target load observation brightness temperature range. By combining the effective calibration samples obtained from multiple natural calibration regions, a joint calibration sample set covering the preset brightness temperature dynamic range of the target load is constructed. The natural calibration region includes stable natural scenes with different brightness temperature levels and different underlying surface types; For example, it includes two or more natural scenes with different microwave radiation characteristics, temporal stability and spatial uniformity, and expands the brightness temperature coverage of the channel to be calibrated in the establishment of the calibration model by the brightness temperature complementarity relationship between different natural scenes.

[0030] like Figure 2 As shown, this embodiment selects the Amazon rainforest region, the equatorial Pacific Ocean region, the Taklamakan Desert region, and the Antarctic ice sheet region as natural calibration regions.

[0031] Among them, the Amazon rainforest region mainly provides high brightness temperature samples, the equatorial Pacific Ocean mainly provides low brightness temperature samples, and the Taklamakan Desert region and the Antarctic ice sheet region provide samples at intermediate brightness temperature levels.

[0032] By combining different natural scenes, the data participating in the cross-calibration covers a wide brightness temperature range from low to high brightness temperature, as shown in Table 1. Table 1. Specific spatial range of each natural calibration region ; The aforementioned multiple natural calibration regions together form a multi-natural scene joint calibration sample system to reduce the problem of extrapolating calibration relationships caused by the limited brightness temperature range of samples when only a single natural scene is used.

[0033] S2. Matching the target channel with the reference channel: For any uncalibrated channel of the target load, the corresponding reference channel is determined from the reference load based on the center frequency, polarization mode and observation parameters of the uncalibrated channel; Establish the correspondence between the target load calibration channel and the reference load reference channel; The matching criteria between the target load calibration channel and the reference load reference channel include: the center frequency being the same or approximately the same, the polarization mode being consistent or corresponding, and the observation angle being the same or corrected by radiation transfer simulation based on the difference in observation angle.

[0034] For each channel of the SMR to be calibrated, a channel with the same or similar center frequency and corresponding polarization is selected from AMSR2 as a reference channel based on the center frequency and polarization mode.

[0035] The frequency matching relationships are shown in Table 2; Table 2 Channel Frequency Matching Relationship ; For each of the above frequencies, a corresponding target channel-reference channel matching relationship is established according to V polarization and H polarization respectively; for channels that are actually set to only one polarization mode, the corresponding relationship is determined according to the actual channel configuration of the load.

[0036] like Figure 3 As shown, for any channel to be calibrated, the target payload and the reference payload are simulated for subsequent radiative transfer according to their respective actual center frequencies, polarization modes, and observation angles. When there is a certain difference between the center frequencies of the target payload and the reference payload, they are not directly assumed to have identical theoretical brightness temperatures; instead, the differences in channel frequency and observation conditions are considered through separate simulations.

[0037] S3. Acquisition and preprocessing of observation data and environmental parameters: Acquire the observed brightness temperature data of the target load and reference load in the corresponding channels of each natural calibration region, and acquire the time information, spatial location information and environmental parameters corresponding to the observed brightness temperature data; The acquired data is subjected to quality control, spatiotemporal matching, and data processing to obtain effective calibration samples. The environmental parameters include parameters describing atmospheric conditions and parameters describing surface conditions; the atmospheric condition parameters include one or more of temperature, humidity, and cloud water content. The surface state parameters include surface temperature, soil moisture content, vegetation parameters, sea surface state parameters, and ice and snow parameters. One or more of the following.

[0038] For example, acquire observational data for HY-2B / SMR and GCOM-W1 / AMSR2 from 2019 to 2020. The observational data includes at least the observed brightness temperature, longitude, latitude, observation time, center frequency, polarization mode, and mass labeling information. Based on the four natural calibration regions determined in step S1, extract valid observation samples falling within the corresponding regions from the target payload and reference payload observational data. Filter the observational data according to preset quality control conditions, removing abnormal brightness temperature data and invalid data affected by rainfall, sea ice, or other non-target factors. Simultaneously, acquire ERA5 reanalysis data corresponding to the satellite observation time and location, which will be used as the environmental parameters required for microwave radiative transfer calculations. For different types of natural calibration regions, select corresponding environmental parameters based on their underlying surface properties, as shown in Table 3. Table 3 Environmental Parameter Sets for Each Region ; Based on the actual observation time and location of the target and reference payloads, satellite observation data are matched temporally and spatially with ERA5 environmental parameters. For data with different spatial resolutions, interpolation or resampling is used to unify them to a preset spatial scale. In this embodiment, a 7-day statistical time window is used to collect effective observed brightness temperatures and their corresponding environmental parameters within the same natural calibration area. Outliers are removed using the 3σ criterion, and weekly average observed brightness temperatures and weekly average environmental parameters are further calculated as input data for subsequent radiative transfer simulation and double-difference calculations.

[0039] S4. Model-assisted simulation of brightness temperature calculation; Based on the underlying surface type corresponding to different natural calibration areas, the corresponding simulated brightness temperature calculation model is selected, and combined with the atmospheric radiation transfer model, the simulated brightness temperature corresponding to the target load calibration channel and the reference load reference channel is calculated respectively. The target load and the reference load are simulated according to their respective actual center frequencies, polarization modes and observation conditions, so that the obtained simulated brightness temperatures correspond to the actual observation states of the target load and the reference load, respectively. The atmospheric radiative transfer model is used to calculate the theoretical simulated brightness temperature of the corresponding channel based on environmental parameters, observation frequency, polarization mode, and observation angle. The simulated brightness temperature calculation model includes an atmospheric radiation transfer model and a surface microwave emissivity model corresponding to different underlying surface types.

[0040] This embodiment uses the RTTOV model to calculate atmospheric microwave radiative transmission. For the equatorial Pacific Ocean scenario, the FASTEM model is used to calculate the microwave emissivity of the ocean surface; for the Amazon rainforest and Taklamakan desert scenarios, the LandEM module in CRTM is used to calculate the microwave emissivity of the land surface; for the Antarctic ice and snow scenario, the SnowEM module in CRTM is used to calculate the microwave emissivity of the ice and snow surface.

[0041] For the same natural calibration sample, SMR and AMSR2 use spatiotemporally matched environmental state parameters for simulation, but input the actual center frequency, polarization mode, and observation angle of the corresponding channel for each of the two loads respectively, thereby obtaining the simulated brightness temperature of the target load. Simulated brightness temperature under reference load Taking the 37 GHz-H polarization channel as an example, the simulated brightness temperature obtained in the Amazon rainforest, equatorial Pacific Ocean, Taklamakan Desert, and Antarctic ice sheet regions is compared with the actual observed brightness temperature. Figure 4 As shown.

[0042] Depend on Figure 4It is evident that within different natural calibration regions, the simulated brightness temperature can reflect the overall variation characteristics of the actual observed brightness temperature, while a certain deviation still exists between the simulated and actual observed brightness temperatures. Therefore, a double-difference deviation is further constructed by utilizing the simulation-observation deviations of the target load and the reference load respectively.

[0043] S5. Calculation of double difference deviation: like Figure 3 As shown, for any channel to be calibrated of the target load, the first deviation between the simulated brightness temperature and the actual observed brightness temperature of the target load channel to be calibrated is calculated, and the second deviation between the simulated brightness temperature and the actual observed brightness temperature of the reference channel of the reference load is calculated; based on the difference between the first deviation and the second deviation, the double difference deviation of the corresponding channel to be calibrated is obtained. Specifically, the first deviation between the simulated brightness temperature of the target load and the actual observed brightness temperature is calculated: ; in, This represents the simulated brightness temperature of the target load channel to be calibrated. This indicates the actual observed brightness temperature of the target payload's uncalibrated channel; Calculate the second deviation between the simulated brightness temperature and the observed brightness temperature of the reference channel for the reference load: ; in, This indicates the simulated brightness temperature of the reference channel of the reference load. This indicates the actual observed brightness temperature of the reference channel of the reference load.

[0044] Construct the double-difference deviation DD between the target load and the reference load: ; Through the above processing, the partial radiative transfer model error and environmental parameter error jointly experienced by the target load and the reference load are reduced in the difference process, thereby obtaining the double difference deviation sample corresponding to the current channel to be calibrated. S6. Establishment of a joint correction model across natural scenes: For the same channel to be calibrated, the double-difference deviation samples obtained from different natural calibration regions are combined; linear regression is performed with the observed brightness temperature of the target load as the independent variable and the corresponding double-difference deviation amount as the dependent variable to establish the linear relationship between the double-difference deviation amount and the observed brightness temperature of the target load, and the correction parameters of the target load channel to be calibrated are obtained. Specifically, based on the original observed brightness temperature of the target payload Using as the independent variable and the corresponding double difference deviation DD as the dependent variable, a linear regression method is used to establish the relationship between the two:

[0045] Where a is the linear correction slope corresponding to the current channel to be calibrated, and b is the linear correction intercept corresponding to the current channel to be calibrated. Taking 37GHz-H as an example, the regression fitting results of the joint samples from the four natural calibration regions are as follows: Figure 5 As shown. Different natural calibration regions provide data samples with different brightness temperature levels, enabling the joint sample to cover a wide range of target payload observation brightness temperatures. The fitting relationship obtained from this representative channel is: ; Following the same method, independent linear regressions were performed on all uncalibrated frequency-polarization channels of the SMR to obtain the double-difference deviation fitting coefficients for each uncalibrated channel, as shown in Table 4 below. Table 4. Fitting coefficients for double difference deviations in each channel ; This results in a set of channel-level calibration coefficients that correspond one-to-one with each of the target load's calibration channels, instead of using a unified calibration coefficient to correct all channels as a whole.

[0046] S7, Channel-level brightness temperature correction: The correction amount corresponding to the target payload's uncalibrated channel is calculated based on the correction mapping relationship; the correction amount is applied to the original observed brightness temperature of the target payload to obtain the corrected brightness temperature after cross-calibration; steps S2 to S7 are executed for different frequency and polarization channels of the target payload to obtain the independent correction parameters corresponding to each uncalibrated channel, thus achieving channel-level cross-calibration of the spaceborne microwave radiometer. Specifically, the brightness temperature correction amount corresponding to the current uncalibrated channel is defined as follows: ; Adding the correction amount to the original observed brightness temperature of the target payload, we obtain the corrected brightness temperature after cross-calibration: ; Right now: ; in, This indicates the corrected brightness temperature of the target load channel after joint double-difference cross-calibration of multiple natural scenes.

[0047] For the SMR channels of different frequencies and polarizations to be calibrated, brightness temperature corrections were calculated using their respective double-difference linear relationship parameters a and b, achieving independent cross-calibration for different frequency-polarization channels. To verify the effectiveness of the above-mentioned multi-natural-scene joint double-difference cross-calibration method, the linear correction coefficients obtained for each channel to be calibrated were applied to the original SMR observed brightness temperature, and the corrected SMR brightness temperature was compared with the observed brightness temperature of the corresponding reference channel of AMSR2.

[0048] Under the same valid sample and statistical conditions, the deviation, standard deviation, root mean square error, and correlation coefficient between the target load and the reference load before and after cross-calibration were calculated. These are shown in Table 5 below.

[0049] Table 5. Calibration Results in the Amazon Rainforest Region ; The average brightness temperature error between the reference load and the cross-calibration in the representative Amazon rainforest region is as follows: Figure 6 As shown, the effectiveness of the established linear correction relationship can be evaluated by comparing the relationship between the brightness temperature observed in the corresponding channel of the target load before and after correction and the reference load.

[0050] When the corrected target load observation brightness temperature exhibits a small systematic bias and dispersion compared to the uncorrected version, and maintains a stable correction effect in different natural scenarios such as the Amazon rainforest, the equatorial Pacific Ocean, the Taklamakan Desert, and the Antarctic ice sheet, it indicates that the obtained channel-level linear correction coefficients can be used for cross-calibration of the corresponding target load channels.

[0051] The above embodiments illustrate that this application combines multiple natural calibration regions with different underlying surface types and different brightness temperature levels to conduct radiation transfer simulations according to the actual channel conditions of the target load and the reference load. By constructing a double-difference deviation through two sets of simulation-observation biases, and by using joint samples across natural scenarios to solve the linear correction relationship of different frequency-polarization channels of the target load, the independent cross-calibration of the spaceborne microwave radiometer at the channel level can be achieved.

[0052] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A channel-level cross-calibration system for a spaceborne microwave radiometer, characterized in that: Including, The natural calibration region determination module is used to select multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics to construct a joint calibration sample set. The channel matching module is used to establish the correspondence between the target load calibration channel and the reference load reference channel; The data acquisition and processing module is used to acquire the observed brightness temperature data and environmental parameters of the target load and reference load, and to perform quality control and spatiotemporal matching processing. The simulated brightness temperature calculation module is used to calculate the simulated brightness temperature of the corresponding channels of the target load and the reference load according to the type of the underlying surface in different natural calibration regions. The double-difference deviation calculation module is used to obtain the double-difference deviation amount based on the difference between the simulation-observation deviation of the target load and the simulation-observation deviation of the reference load. The calibration model establishment module is used to combine double-difference bias samples from multiple natural calibration regions to establish a linear calibration relationship between the double-difference bias of each channel to be calibrated and the observed brightness temperature. The brightness temperature correction module is used to calculate the correction amount according to the linear correction relationship and correct the original observed brightness temperature of the target load to obtain the corrected brightness temperature after independent calibration of each channel.

2. A method for cross-calibrating a channel-level spaceborne microwave radiometer using the channel-level cross-calibration system as described in claim 1, characterized in that: Multiple geographically separated natural calibration regions with different underlying surface types and complementary brightness temperature characteristics were selected to construct joint calibration samples covering different brightness temperature ranges; for the target load to be calibrated channel, the corresponding reference channel was determined from the reference load. The observed brightness temperature and environmental parameters of the corresponding channels of the target payload and reference payload are obtained. Combined with the surface microwave emissivity model and atmospheric radiative transfer model of different natural scenarios, the simulated brightness temperature of the corresponding channels of the target payload and reference payload is calculated respectively. Based on the deviation between the simulated brightness temperature and the actual observed brightness temperature, a double-difference deviation is constructed. By combining multiple natural scenario samples, a channel-level correction relationship between the double-difference deviation and the observed brightness temperature of the target payload is established. The correction parameters corresponding to each frequency-polarization channel are obtained independently, and the observed brightness temperature of the target payload is corrected.

3. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 2, characterized in that: Different frequency-polarization channels of the target load have independent corresponding linear correction slopes and linear correction intercepts, and different channels do not share the same correction parameters.

4. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 2, characterized in that: The implementation process includes the following steps: S1. Selection of natural calibration scenarios and construction of joint calibration samples; Multiple geographically separated natural calibration regions with complementary microwave radiation brightness temperature characteristics are selected. The temporal stability and spatial uniformity of each natural calibration region are screened to ensure that each natural calibration region provides effective calibration samples for different brightness temperature ranges within the target load observation brightness temperature range. The effective calibration samples obtained from multiple natural calibration regions are combined to construct a joint calibration sample set covering the preset brightness temperature dynamic range of the target load. S2. Matching the target channel with the reference channel: For any uncalibrated channel of the target load, the corresponding reference channel is determined from the reference load based on the center frequency, polarization mode and observation parameters of the uncalibrated channel; Establish the correspondence between the target load calibration channel and the reference load reference channel; S3. Acquisition and preprocessing of observation data and environmental parameters: Acquire the observed brightness temperature data of the target payload and reference payload in the corresponding channels of each natural calibration region, and acquire the time information, spatial location information and environmental parameters corresponding to the observed brightness temperature data; perform quality control, spatiotemporal matching and data processing on the acquired data to obtain effective calibration samples; S4. Model-assisted simulation of brightness temperature calculation; Based on the underlying surface type corresponding to different natural calibration areas, the corresponding simulated brightness temperature calculation model is selected, and combined with the atmospheric radiation transfer model, the simulated brightness temperature corresponding to the target load calibration channel and the reference load reference channel is calculated respectively; the target load and the reference load are simulated according to their respective actual center frequency, polarization mode and observation conditions, so that the obtained simulated brightness temperatures correspond to the actual observation states of the target load and the reference load respectively. S5. Calculation of double difference deviation: For any channel to be calibrated of the target load, calculate the first deviation between the simulated brightness temperature and the actual observed brightness temperature of the target load channel to be calibrated, and the second deviation between the simulated brightness temperature and the actual observed brightness temperature of the reference channel of the reference load; based on the difference between the first deviation and the second deviation, obtain the double difference deviation of the corresponding channel to be calibrated. Calculate the first deviation between the simulated brightness temperature and the actual observed brightness temperature of the target load: ； in, This represents the simulated brightness temperature of the target load channel to be calibrated. This indicates the actual observed brightness temperature of the target payload's uncalibrated channel; Calculate the second deviation between the simulated brightness temperature and the observed brightness temperature of the reference channel for the reference load: ； in, This indicates the simulated brightness temperature of the reference channel of the reference load. This indicates the actual observed brightness temperature of the reference channel of the reference load; Construct the double-difference deviation DD between the target load and the reference load: ； Through the above processing, the partial radiative transfer model error and environmental parameter error jointly experienced by the target load and the reference load are reduced in the difference process, thereby obtaining the double difference deviation sample corresponding to the current channel to be calibrated. S6. Establishment of a joint correction model across natural scenes: For the same channel to be calibrated, the double-difference deviation samples obtained from different natural calibration regions are combined; linear regression is performed with the observed brightness temperature of the target load as the independent variable and the corresponding double-difference deviation amount as the dependent variable to establish the linear relationship between the double-difference deviation amount and the observed brightness temperature of the target load, and the correction parameters of the target load channel to be calibrated are obtained. S7, Channel-level brightness temperature correction: Calculate the correction amount corresponding to the target load's uncalibrated channel based on the aforementioned correction mapping relationship; The correction amount is applied to the original observed brightness temperature of the target payload to obtain the corrected brightness temperature after cross-calibration. Steps S2 to S7 are executed for different frequencies and polarization channels of the target payload to obtain the independent calibration parameters for each channel to be calibrated, thereby achieving channel-level cross-calibration of the spaceborne microwave radiometer.

5. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 4, characterized in that: The S1 mentioned above refers to a natural calibration region that includes two or more natural scenes with different microwave radiation characteristics, temporal stability and spatial uniformity. The brightness temperature coverage of the channel to be calibrated is expanded by the brightness temperature complementarity between different natural scenes.

6. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 5, characterized in that: The matching criteria for the target load calibration channel and the reference load reference channel in S2 include: the center frequency being the same or approximately the same, the polarization mode being consistent or corresponding, and the observation angle being the same or corrected by radiation transfer simulation based on the difference in observation angle.

7. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 6, characterized in that: The aforementioned S3, Environmental parameters include parameters that describe the state of the atmosphere and parameters that describe the state of the Earth's surface. The atmospheric state parameters include one or more of temperature, humidity, and cloud liquid water content; The surface condition parameters include one or more of the following: surface temperature, soil moisture content, vegetation parameters, sea surface condition parameters, and ice and snow parameters.

8. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 7, characterized in that: The S4 atmospheric radiative transfer model is used to calculate the theoretical simulated brightness temperature of the corresponding channel based on environmental parameters, observation frequency, polarization mode, and observation angle. The simulated brightness temperature calculation model includes an atmospheric radiation transfer model and a surface microwave emissivity model corresponding to different underlying surface types.

9. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 8, characterized in that: Step S6 involves using the original observed brightness temperature of the target payload. Using as the independent variable and the corresponding double difference deviation DD as the dependent variable, a linear regression method is used to establish the relationship between the two: ； Where a is the linear correction slope corresponding to the current channel to be calibrated, and b is the linear correction intercept corresponding to the current channel to be calibrated.

10. The channel-level cross-calibration method for a spaceborne microwave radiometer according to claim 9, characterized in that: S7, as described above, specifies the brightness temperature correction amount corresponding to the current channel to be calibrated: ； Adding the correction amount to the original observed brightness temperature of the target payload, we obtain the corrected brightness temperature after cross-calibration: ； Right now: ; in, This indicates the corrected brightness temperature of the target load channel after joint double-difference cross-calibration of multiple natural scenes.

Images