System deviation correction method for improving salinity inversion precision of interference type microwave radiometer

Abstract

The invention relates to a system deviation correction method for improving the salinity inversion precision of an interference type microwave radiometer, and relates to the technical field of satellite ocean remote sensing data processing. The method comprises the following steps of: acquiring an area, strictly purifying and screening observed brightness temperature data of a preset time window in the area to eliminate the influence of radio frequency interference, sun and moon pollution and the like, acquiring high-purity original data, calculating simulated brightness temperature by utilizing a radiation transmission model, and subtracting the simulated brightness temperature from the purified observed brightness temperature to obtain a brightness temperature deviation sample; and then a robust statistical method based on a sliding window and a median is adopted to carry out statistical calculation on the deviation sample to obtain a stable and reliable deviation correction value, and finally a two-dimensional lookup table is generated and applied to sea surface salinity inversion. According to the method, the observation brightness temperature for salinity inversion can be efficiently and accurately corrected from the source, and the contradiction between stability and adaptability in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of satellite ocean remote sensing data processing, and particularly relates to a system bias correction method for improving the salinity retrieval accuracy of an interferometric microwave radiometer. BACKGROUND

[0002] Sea Surface Salinity (SSS) is a key physical parameter affecting ocean dynamic processes and global thermohaline circulation. Many phenomena and processes occurring in the ocean are closely related to its distribution and variation. Therefore, studying the distribution and variation of sea surface salinity has irreplaceable application value for improving the prediction ability of ocean and climate models, monitoring extreme weather events, and conducting ocean science research.

[0003] Satellite remote sensing has become the core support for monitoring sea surface salinity. Compared with traditional in-situ measurements relying on ships and buoys, satellite microwave remote sensing technology, especially L-band (about 1.4 GHz) microwave radiometers, can penetrate clouds and achieve all-weather, large-scale, and high-frequency global sea surface salinity observations.

[0004] However, there is a fundamental technical challenge in using microwave radiometers to retrieve sea surface salinity: the measured basic physical quantity, brightness temperature (TB), has very low sensitivity to salinity changes. For every 1 practical salinity unit (PSU) change in salinity, the brightness temperature changes only about 0.5 K. This weak signal response makes the observed brightness temperature easily disturbed by noise from multiple sources, and any small bias can be amplified into a huge salinity error, thus drowning the real ocean salinity signal. These biases mainly come from the uncertainty of the geophysical model and the pollution of external environmental signals. Currently, existing bias correction schemes either use sparse in-situ data for large-scale statistical correction after the generation of salinity products, failing to address the quality problem of input brightness temperature from the source, or rely on complex numerical weather prediction (NWP) data assimilation systems, which are costly and difficult to deploy independently. These methods are difficult to balance stability and adaptability to time-varying biases, and the reference benchmark used for statistical bias is easily contaminated by complex global sea conditions and noise, resulting in insufficient correction accuracy and robustness. Therefore, before salinity retrieval, using a high-precision, high-stability, and computationally efficient method to correct the bias of the original observed brightness temperature has great technical significance and application value for improving the overall accuracy and reliability of sea surface salinity products. SUMMARY

[0005] The present application aims to solve the above problems existing in the prior art, and provides a brightness temperature deviation correction method based on a stable reference area and robust statistics, aiming to provide a high-precision, high-stability and high-efficiency solution. The core of the present application is to establish a stable and pure deviation correction reference, and to generate a quasi-dynamic correction lookup table through robust statistical means.

[0006] The present application provides a system deviation correction method for improving the salinity inversion accuracy of an interferometric microwave radiometer, comprising the following steps: S1. Selecting a stable calibration reference area for the interferometric microwave radiometer; S2. Collecting observed brightness temperature data in a preset time window in the reference area; S3. Purifying and selecting the observed brightness temperature data to obtain an effective data set; S4. Calculating the simulated brightness temperature corresponding to each observation point in the effective data set using a microwave radiation transmission model, and subtracting the observed brightness temperature to obtain brightness temperature deviation samples; S5. Statistically calculating the deviation samples to obtain the final deviation correction value; S6. Storing the final deviation correction value of S5 in a two-dimensional lookup table; S7. When performing sea surface salinity inversion, obtaining the final deviation correction value from the two-dimensional lookup table, correcting the observed brightness temperature, and then performing sea surface salinity inversion.

[0007] Preferably, in step S1, the selection of the stable calibration reference area is based on the following four conditions: The marine dynamic environment is stable, and the temporal and spatial variations of sea surface temperature and salinity are gentle; Far away from land and human activity area, with low risk of radio frequency interference and land pollution; Uniform weather conditions; The shape of the region is similar to the satellite observation orbit path.

[0008] Preferably, in step S2, the preset time window is a sliding time window centered on the target correction day, and the sliding time window covers a total of 10 days of data period.

[0009] Preferably, in step S5, the geographical coordinates are converted to central observation direction cosine coordinates using the observation point information, and mapped to two-dimensional indices; in the sliding time window, for each index, the median of all brightness temperature deviation samples corresponding to the index is calculated, and the median is taken as the final deviation correction value.

[0010] Preferably, the geographical coordinates of the center point of the instrument field of view are discretized into The index comprises the following steps: first, based on the real-time position and velocity vector of the satellite and the pointing angle of the instrument, establishing the satellite platform coordinate system and the instrument coordinate system as the reference; then, converting the geographic coordinates of the field of view center point into the earth-centered inertial coordinate system and into the instrument coordinate system to obtain the field of view direction vector; subsequently, calculating the direction cosine of the field of view direction vector in the instrument coordinate system; finally, through normalization processing, mapping the direction cosine into the preset grid Index.

[0011] Preferably, the step of establishing the satellite platform coordinate system and the instrument coordinate system comprises the following steps: using the satellite position vector and the velocity vector to define the platform coordinate basis vector , , :

[0012]

[0013]

[0014] Using the instrument pitch angle and the rotation angle to generate the instrument coordinate system rotation matrix : .

[0015] Preferably, the step of converting the geographic coordinates of the field of view center point into the instrument coordinate system comprises the following steps: first, converting the ground point with the latitude and the longitude into the ECEF coordinate vector :

[0016] wherein R is the radius of the earth; then obtaining the field of view direction vector in the instrument coordinate system through the following formula :

[0017] , and are the projection components of the field of view direction vector on the x-axis, the y-axis and the z-axis of the instrument coordinate system, respectively.

[0018] Preferably, the step of calculating the direction cosine comprises the following steps:

[0019]

[0020] wherein, , , denotes the cosine of the angle between the x-axis of the instrument coordinate system, denotes the cosine of the angle between the y-axis of the instrument coordinate system, is the zenith angle, representing the angle between the positive direction of the z-axis of the instrument coordinate system, is the azimuth angle, representing the angle between the projection on the x-y plane and the positive direction of the x-axis.

[0021] Preferably, the direction cosine is mapped to the index The step is specifically: first, determine the maximum value (cos max, cos min) of the direction cosine in the calibration reference area, and generate a 129x129 standard grid in the range to store the final bias correction value; , , The index is calculated by the following formula: ; .

[0022] Preferably, in the application stage of the sea surface salinity inversion in the step S7, the direction cosine coordinate of each observation point needs to be calculated first; then, the two-dimensional lookup table is linearly interpolated through the coordinate to accurately obtain the final bias correction value of the coordinate position. Subtract the interpolation result from the original observation brightness temperature to obtain the brightness temperature data corrected by the system bias.

[0023] The beneficial effects of the present application are: The purpose of the present application is to solve the problems in the prior art, and a high-precision and high-stability brightness temperature bias correction method for L-band microwave radiometer is proposed. The specific calm sea area is used as a natural calibration field, and a quasi-dynamic bias correction lookup table is generated through an innovative calibration scheme based on stable reference area and robust statistics, so that the observed brightness temperature is efficiently, accurately and stably corrected, thereby significantly improving the precision and reliability of the sea surface salinity inversion product. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is the flowchart of the brightness temperature bias correction method for improving the longitude of the sea surface salinity inversion of the present application; Figure 2 is a bias correction value diagram of the present application; ​​Figure 3 This is a schematic diagram illustrating the correction effect of the present invention. Detailed Implementation

[0025] To further understand the present invention, it will be further described below with reference to the accompanying drawings and embodiments, and the technical solutions of the present invention will be clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments. This embodiment proposes a brightness temperature deviation correction method based on a stable reference zone and robust statistics, and applies it to sea surface salinity inversion to verify its effectiveness.

[0027] like Figure 1 The diagram shown is a schematic flowchart of the brightness temperature deviation correction method for improving longitude inversion based on sea surface salinity according to the present invention, which includes the following steps: S1. Select a stable calibration reference area for the interferometric microwave radiometer. The brightness temperature data used in this embodiment is the SMOSL1C grade product, and the auxiliary geophysical parameters (such as sea surface temperature, wind field, etc.) are from the ECMWF auxiliary dataset released in conjunction with SMOS data.

[0028] Before the experiment began, a large and relatively calm quadrilateral area was selected near the center of the South Pacific subtropical circulation as the reference area for dynamic calibration, based on the four conditions for the stable calibration reference area proposed in this invention.

[0029] The selection of a stable calibration reference area is based on the following four conditions: The marine dynamic environment is stable, and the temporal and spatial variations of sea surface temperature and salinity are gradual. Located far from land and areas of human activity, with low risk of radio frequency interference and land pollution; Weather conditions are uniform; The shape of the region is similar to the orbital path of satellite observation.

[0030] S2. Collect the observed brightness temperature data within the preset time window in the reference area and perform data preprocessing.

[0031] This step aims to collect the original data in the reference area, and the preset time window is a sliding time window centered on the target correction day. Specifically, a sliding time window of 10 days (D-5 to D+5) is constructed with any target correction day (D) as the center. All SMOS L1C brightness temperature observation data in the calibration reference area within this window are collected.

[0032] S3. Purify and select the observed brightness temperature data to obtain an effective data set.

[0033] Any geographic point that does not meet the quality requirements is removed, and the removal conditions include: 1) external signal interference: radio frequency interference, sun or moon pollution; 2) severe geophysical conditions: rain, sea ice coverage, extreme wind speed (too high or too low); 3) data itself defects: instrument error, missing key auxiliary data, or insufficient amount of effective observed brightness temperature data.

[0034] After the above screening, a high-purity original data set is obtained, denoted as data set A.

[0035] S4. Calculate the simulated brightness temperature corresponding to each observation point in the effective data set using the microwave radiation transfer model, and subtract the observed brightness temperature to obtain the brightness temperature bias ΔTB sample.

[0036] Using the radiation transfer model, the simulated brightness temperature corresponding to each effective observation point in data set A is calculated. Then, the observed brightness temperature is subtracted from the simulated brightness temperature to obtain the brightness temperature bias ΔTB sample of the point.

[0037] S5. Robust statistics are performed on the bias samples to obtain the final bias correction value.

[0038] To ensure the stability and reliability of the correction value, a robust statistical method is used in this step.

[0039] The geographic coordinates are converted to central observation direction cosine coordinates using the observation point information, and are mapped to two-dimensional index; in the sliding time window, for each index, the median of all brightness temperature bias samples corresponding to the index is calculated, and the median is taken as the final bias correction value.

[0040] The geographic coordinates of the center point of the instrument field of view are discretized into index, including the following steps: first, based on the real-time position and velocity vector of the satellite and the pointing angle of the instrument, the satellite platform coordinate system and the instrument coordinate system are established as the reference; then, the geographic coordinates of the center point of the field of view are converted to the geocentric coordinate system and to the instrument coordinate system, obtaining the field of view direction vector; then, according to the direction vector, the direction cosine in the instrument coordinate system is calculated; finally, through normalization processing, the direction cosine is mapped to the index.

[0041] The specific steps for establishing the satellite platform coordinate system and the instrument coordinate system are as follows: using the satellite position vector and velocity vector Define platform coordinate basis vector , , :

[0042]

[0043]

[0044] Using the instrument's pitch angle Rotation angle Generate the rotation matrix of the instrument coordinate system : .

[0045] The specific steps for converting the geographic coordinates of the field of view center point to the instrument coordinate system are as follows: First, convert the latitude... Longitude is Converting ground points to ECEF coordinate vectors :

[0046] Where R is the Earth's radius; The field of view direction vector in the instrument coordinate system is then obtained using the following formula. That is, the direction vector from the satellite to the center point of the observation field of view:

[0047] , and These are the field of view direction vectors. Projected components on the x-axis, y-axis, and z-axis of the instrument coordinate system.

[0048] Calculate direction cosine The specific steps are as follows:

[0049]

[0050] in, express The cosine of the angle between the instrument coordinate system and the x-axis. express The cosine of the angle between the instrument coordinate system and the y-axis; The physical meaning is the straight-line distance between the satellite and the center point of the observation field of view on the ground. is the zenith angle, which represents the angle between the positive direction of the z-axis of the instrument coordinate system and the straight line connecting the satellite and the center point of the observation field of view on the ground, is the azimuth angle, which represents the angle between the projection of the straight line connecting the satellite and the center point of the observation field of view on the ground on the x-y plane and the positive direction of the x-axis, which can be calculated by the formula , .

[0051] The direction cosine is mapped to the index The specific steps are as follows: first, determine the maximum value (θmax, φmax) and the minimum value (θmin, φmin) of the direction cosine in the calibration reference area, and generate a standard grid of 129x129 in this range to store the final bias correction value; , , The index is calculated by the following formula: ; .

[0052] Specifically, all ΔTB samples obtained in a 10-day sliding window are mapped to a regular two-dimensional (x, y) grid through direction cosine transformation. Then, the final bias correction value of the target day (D) is determined by calculating the median of all samples in each grid cell. The final bias correction value is shown in Figure 2 .

[0053] S6. Store the final bias correction value in a two-dimensional lookup table Store the calculated bias correction value in a two-dimensional lookup table with the grid cell in step 5 as the index.

[0054] S7. When performing sea surface salinity retrieval, obtain the final bias correction value from the two-dimensional lookup table, correct the observed brightness temperature, and then perform sea surface salinity retrieval.

[0055] Apply the lookup table generated in step S6 to the global sea surface salinity retrieval process to verify its correction effect. Before performing salinity retrieval on any observed brightness temperature, first interpolate the corresponding bias correction value from the lookup table, and subtract the value from the observed brightness temperature. Use the corrected brightness temperature as input to perform sea surface salinity retrieval, thereby obtaining a more accurate sea surface salinity value.

[0056] As Figure 3 ​​As shown, (a) is the sea surface salinity map retrieved using original brightness temperature data, (b) is the sea surface salinity map retrieved after bias correction by applying the method of the present application, and (c) is the difference between the sea surface salinity before and after correction. By comparison, it can be seen that the structure of the sea surface salinity field after correction is clearer in the main ocean current area, and the large-scale artifacts and biases existing in the original are eliminated, and it is more consistent with known oceanographic characteristics, proving the effectiveness of the method of the present application.

[0057] The above description is merely preferred embodiments of the present application, but not a limitation of the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art will still be able to make modifications to the technical solutions described in the foregoing embodiments, or make equivalent replacements to some of the technical features. Any modifications, equivalent replacements, modifications, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for correcting systematic biases to improve the accuracy of salinity inversion in interferometric microwave radiometers, characterized in that, Includes the following steps: S1. Select a stable calibration reference area for the interferometric microwave radiometer; S2. Collect the observed brightness temperature data within the preset time window of the reference area; S3. Clean and filter the observed brightness temperature data to obtain a valid dataset; S4. Calculate the simulated brightness temperature corresponding to each observation point in the effective data using the microwave radiative transfer model and subtract it from the observed brightness temperature to obtain the brightness temperature deviation sample. S5. Perform statistical calculations on the deviation samples to obtain the final deviation correction value; S6. Store the final deviation correction value of S5 into a two-dimensional lookup table; S7. When performing sea surface salinity inversion, obtain the final deviation correction value from the two-dimensional lookup table, correct the observed brightness temperature, and then perform sea surface salinity inversion.

2. The method for correcting system bias to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 1, characterized in that, In step S1, the selection of the stable calibration reference area is based on the following four conditions: The marine dynamic environment is stable, and the temporal and spatial variations of sea surface temperature and salinity are gradual. Located far from land and areas of human activity, with low risk of radio frequency interference and land pollution; Weather conditions are uniform; The shape of the region is similar to the orbital path of satellite observation.

3. The method for correcting systematic biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 1, characterized in that, In step S2, the preset time window is a sliding time window centered on the target correction date, and the sliding time window covers a total data period of 10 days.

4. The method for correcting system bias to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 3, characterized in that, In step S5, the geographic coordinates are converted into cosine coordinates of the center observation direction using the observation point information, and then mapped to two dimensions. Index; within the sliding time window, for each The index is used to calculate the median of all corresponding brightness temperature deviation samples, and the median is used as the final deviation correction value.

5. The method for correcting system bias to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 4, characterized in that, Discretize the geographic coordinates of the instrument's field of view center point as follows: The indexing process includes the following steps: First, based on the satellite's real-time position and velocity vectors and the instrument's pointing angle, a satellite platform coordinate system and an instrument coordinate system are established as references. Next, the geographic coordinates of the field of view's center point are converted to geocentric coordinates and then transformed to the instrument coordinate system to obtain the field of view direction vector. Subsequently, the direction cosine of this direction vector in the instrument coordinate system is calculated. Finally, through normalization, the direction cosine is mapped to a preset grid. index.

6. The method for correcting system biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 5, characterized in that, The specific steps for establishing the satellite platform coordinate system and the instrument coordinate system are as follows: using the satellite position vector and velocity vector Define platform coordinate basis vector , , : Using the instrument's pitch angle Rotation angle Generate the rotation matrix of the instrument coordinate system : 。 7. The method for correcting system biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 5, characterized in that, The specific steps for converting the geographic coordinates of the field of view center point to the instrument coordinate system are as follows: First, convert the latitude... Longitude is Converting ground points to ECEF coordinate vectors : Where R is the Earth's radius; The field of view direction vector in the instrument coordinate system is then obtained using the following formula. : , and These are the field of view direction vectors. Projected components on the x-axis, y-axis, and z-axis of the instrument coordinate system.

8. The method for correcting system biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 5, characterized in that, The specific steps for calculating the direction cosine are as follows: in, , , express The cosine of the angle between the instrument coordinate system and the x-axis. express The cosine of the angle between the instrument coordinate system and the y-axis. Zenith angle, representing The angle between the instrument coordinate system and the positive z-axis. Azimuth, representing The angle between the projection onto the xy plane and the positive x-axis.

9. The method for correcting system biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 5, characterized in that, Map direction cosine to index The specific steps are as follows: First, determine the maximum value of the direction cosine within the calibration reference area ( , ) and minimum value ( , Within this range, a standard grid of 129×129 is generated to store the final deviation correction value; The index is calculated using the following formula: ； 。 10. The method for correcting system biases to improve the accuracy of salinity inversion in an interferometric microwave radiometer according to claim 1, characterized in that, In step S7, during the application stage of sea surface salinity inversion, the direction cosine coordinates of each observation point need to be calculated first. Then, the two-dimensional lookup table is linearly interpolated using the direction cosine coordinates to accurately obtain the final deviation correction value of the coordinate position. This interpolation result is subtracted from the original observed brightness temperature to obtain the brightness temperature data after systematic deviation correction.

Images