Ocean two-dimensional flow field joint construction method based on SST and SAR

Abstract

The invention relates to the technical field of ocean observation and environment monitoring, in particular to an ocean two-dimensional flow field joint construction method based on SST and SAR, and the method comprises the steps: firstly, obtaining SST data and SAR data which are matched in time and space; then, the SST data are calculated through an MCC algorithm, the ocean surface temperature change is obtained, and the flow direction angle of the ocean flow field is obtained through inversion; sAR data are calculated through a DCA algorithm, and the radial speed of the ocean flow field is obtained through inversion; and finally, combining the flow direction angle and the radial speed of the ocean flow field through a vector synthesis method, and inverting a two-dimensional flow field vector of the ocean flow field. According to the method, the SST data and the SAR data are effectively combined and complemented, the limitation of a single remote sensing means is overcome, the unique advantages of different types of remote sensing data are fully utilized, and the spatial resolution and precision of an inversion result are improved; and the blank of sub-mesoscale ocean current data is effectively filled.

Description

Technical Field

[0001] The present invention relates to the technical field of ocean observation and environmental monitoring, and in particular to a method for jointly constructing an ocean two-dimensional flow field based on SST and SAR. Background Art

[0002] Ocean currents, defined as large-scale movements of seawater with relatively stable velocity and direction, are a crucial component of global ocean circulation. The movement and dynamics of ocean currents not only significantly influence global and regional climate change but are also key factors in the evolution of numerous ocean phenomena and dynamic processes, such as cold and warm water masses, mesoscale eddies, upwellings, boundary currents, and coastal currents. Accurately inverting ocean currents is crucial for monitoring environmental changes, maintaining maritime safety, and supporting shipping, fisheries, and offshore exploration.

[0003] Currently, the commonly used methods for observing ocean currents include in situ observations and satellite remote sensing. In situ observations involve placing observation instruments such as ADCP (Acoustic Doppler Current Profiler), XBT (Expendable Bathythermograph), or buoys in the ocean environment to measure ocean currents in real time and continuously. While in situ observations can provide highly accurate data, their limited spatial coverage makes it difficult to fully understand the spatiotemporal variations of ocean currents, and the observation costs are high. Compared to in situ observations, satellite remote sensing has the advantages of wide coverage, repeated observations, and high temporal resolution, making it a cost-effective method for observing global ocean currents.

[0004] Sea surface height (SSH) measured by satellite altimeters can be used to estimate geostrophic currents based on geostrophic balance theory. Generally, geostrophic currents derived from bottom-view altimeter data are relatively accurate and exhibit significant advantages in retrieving complex flow fields such as mesoscale circulations and eddies. Satellite altimetry data are used to observe changes in mesoscale ocean currents and explore trends in global and regional mesoscale eddies. The launch of the Surface Water and Ocean Topography (SWOT) altimeter has filled the gap in submesoscale ocean altimetry data, providing new possibilities for retrieving submesoscale ocean currents. However, the classical geostrophic balance theory ignores the influence of centrifugal force on real ocean motion, making it difficult to accurately identify small-scale ocean currents, particularly those with scales less than 100 kilometers. Based on geostrophic balance theory, submesoscale geostrophic currents observed using altimeters exhibit certain errors. Therefore, how to effectively distinguish and separate non-geostrophic currents from the inverted ocean currents to accurately extract geostrophic current information remains a key issue that needs to be addressed.

[0005] Sea surface temperature (SST) is a physical parameter with high temporal and spatial resolution. SST has long been used as a tracer to infer ocean currents using multi-temporal tracer measurement methods, such as the maximum cross-correlation method (MCC). Numerous studies have demonstrated that the MCC method can infer ocean currents from time-series satellite imagery and is suitable for inverting small-scale ocean currents, demonstrating its versatility. Color images from the Geostationary Ocean Color Imager (GOCI) were used to calculate ocean currents, revealing the path of the Kuroshio Current. However, the accuracy of this method may be affected in areas with small vector gradients.

[0006] Synthetic aperture radar (SAR) can invert ocean currents through advanced inversion methods such as Doppler centroid shift (DCA) and along-track interferometry (ATI), thereby capturing more accurate and detailed ocean current characteristics. Prior art proposes an improved Doppler radar imaging model (IDopRIM) to improve the accuracy of ocean surface current (OSC) inversion, with a root mean square (RMS) error typically less than 0.3 m / s. Prior art also proposes improving SAR inversion of ocean currents by establishing a geophysical model function (GMF), with a deviation of 0.01 m / s and an RMS error of 0.19 m / s. However, this is limited in that a single SAR image can only invert radial velocity in the radar line of sight and cannot directly obtain a complete two-dimensional flow field. Summary of the Invention

[0007] In response to the above-mentioned defects, the purpose of the present invention is to provide a method for jointly constructing the ocean two-dimensional flow field based on SST and SAR, so as to solve the limitations of a single remote sensing method and the problem of insufficient spatial resolution of the inversion results and insufficient accuracy of flow field construction due to missing processing data.

[0008] To achieve the above objectives, the present invention provides a method for jointly constructing a two-dimensional ocean flow field based on SST and SAR. First, SST data and SAR data that are time-space matched are obtained; Then, the SST data is calculated using the MCC algorithm to obtain the change in ocean surface temperature and invert the flow angle of the ocean current field; the SAR data is calculated using the DCA algorithm to invert the radial velocity of the ocean current field; Finally, the flow direction angle and radial velocity of the ocean flow field are combined through the vector synthesis method to invert the two-dimensional flow field vector of the ocean flow field.

[0009] As a preferred technical solution, the process of obtaining time-space matched SST data and SAR data is as follows: read two images and images The sea surface temperature data is filtered out from the selected sea surface temperature data to obtain the SST data within the required longitude and latitude range and save it.

[0010] As a preferred technical solution, SST data is obtained through the remote sensing satellite Himawari-8, and SAR data is obtained through the remote sensing satellite Sentinel-1. The SST data and SAR data must meet the following requirements: the spatial coverage range overlaps, and the time interval does not exceed 3 hours.

[0011] As an optimal technical solution, the calculation process of SST data is as follows: use the MCC algorithm to calculate the sea temperature gradient and variance within the window, and determine whether the initial window meets the flow field velocity calculation requirements; if not, dynamically adjust the window size, and combine the MCC algorithm to recalculate the sea temperature and flow velocity in the dynamic partition.

[0012] As a preferred technical solution, the calculation process of SST data includes the following steps: Step 11: For the two input images and images Perform mean removal processing and calculate the mean and , get a new image and new images , the formula is as follows: ; Step 12: Calculate the new image and new images The cross-correlation coefficient , used to measure the similarity between images, the formula is as follows: ; By maximizing the cross-correlation coefficient , obtain a preliminary estimate of the velocity field; Step 13: Use the quadratic interpolation method to calculate the cross-correlation matrix The fitting formula is as follows: ; Step 14: Use the neighborhood average method to calculate the velocity vector , identify and correct outliers; for each velocity vector The calculation formula is as follows: ; Among them, N represents the correlation coefficient selected from its spatial neighborhood when calculating the velocity vector of the target position. ≥R th The total number of all valid velocity vectors; where R th =0.65 is the preset cross-correlation coefficient threshold; are two images in the neighborhood and images Velocity vector inside; Outliers include noise and misestimation; Step 15: Use sliding window technology to select square windows of different sizes (such as 16×16 pixels, 32×32 pixels, or 64×64 pixels) for the image. and images Processing; for each position in the image, using the windows of different sizes to respectively calculate the velocity vector of the position to obtain multi-scale velocity field data; Repeat steps 11 to 14 for each window to extract flow features at each scale and generate velocity field vectors. ; Step 16, save the velocity field vector .

[0013] As a preferred technical solution, the steps for extracting radial velocity from SAR data are as follows: Step 21, reading and cleaning SAR data; collecting the cleaned SAR data for multiple days, obtaining multi-date SAR images and splicing and integrating them to form a series of continuous SAR spatiotemporal data; Step 22, extracting backscatter coefficients from SAR spatiotemporal data and performing geometric correction; Step 23: Combine the external wind field ECMWF data, remove samples with wind speeds greater than 10 m / s, and obtain the uniform wind speed and wind direction distribution of the corresponding area of the SAR image through an interpolation method. The interpolation method is: ; in, is the interpolated velocity, is the weight, is the original wind speed at the measurement point; Step 24: Calculate the radial velocity using the following formula based on the geometric information measured by SAR and the sea surface fluctuations: ; in, is the radial flow velocity, is the SAR wavelength, is the angle of incidence, is the Doppler shift; Step 25: Use the CDOP model to correct the calculation results of the radial flow velocity. The core formula of the CDOP model is: ; in, is the corrected speed, is the correction value affected by wind field and incident angle.

[0014] As a preferred technical solution, the result obtained by inverting the SST data using the MCC method is determined as the true flow direction angle, and the result obtained by inverting the SAR data using the DCA method is determined as the true flow velocity. The true flow velocity is mapped to the true direction, thereby inverting the accurate two-dimensional ocean flow field.

[0015] As a preferred technical solution, the steps for constructing the ocean two-dimensional flow field are as follows: Step 31, the flow direction angle extracted from the SST data by the MCC algorithm and the radial velocity results extracted from the SAR data by the DCA algorithm are resampled to the same size grid through Gaussian filtering and bicubic interpolation method; Step 32: Based on the physical relationship between the components of the ocean current, derive the relevant formulas, use formulas (1) and (2) to calculate the meridional velocity and zonal velocity of the two-dimensional flow field of the direct back projection method, and use formula (3) to calculate the meridional velocity and zonal velocity of the two-dimensional flow field of the vector decomposition joint method; (1) (2) (3) In the above formula, is the total vector of the two-dimensional flow field, including the flow velocity (unit: m / s) and the flow direction angle (unit: °, with due north as 0° and increasing clockwise); is the meridional vector of the two-dimensional flow field, is the latitudinal vector of the two-dimensional flow field, is the radial velocity vector, is the flow angle, is the radar viewing angle; In step 33, the difference in the modulus lengths of the two velocity vectors is calculated according to the following formula: (4) in, ; It is the two-dimensional flow field vector inverted by direct back-projection method. It is the two-dimensional flow field vector inverted by the vector decomposition joint method. and represent the longitudinal velocity and latitudinal velocity of the two-dimensional flow field obtained based on the direct backprojection method, and They represent the meridional velocity and zonal velocity of the two-dimensional flow field obtained based on the vector decomposition joint method; Step 34, calculating the correction factor; The correction factor is calculated as follows: (5) in, Express the average velocity difference across all calculation points: (6) Step 35, correcting the velocity component; The second set of velocity vectors is weighted using the correction factor to obtain the optimized velocity vector: (7).

[0016] This paper provides a method for jointly constructing a two-dimensional ocean current field based on SST and SAR. This method combines sea surface temperature (SST) data from the Himawari-8 satellite and synthetic aperture radar (SAR) data from the Sentinel-1 satellite, using a joint inversion model to calculate the two-dimensional current field. This effective combination and complementary nature of SST and SAR not only overcomes the limitations of single remote sensing methods but also leverages the unique advantages of different types of remote sensing data, improving the spatial resolution and accuracy of the inversion results. The average correlation coefficient for inversion accuracy can reach 0.72, and the angular phase difference can reach 5.57°. This method effectively fills the gap in sub-mesoscale ocean current data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a process for jointly constructing a two-dimensional ocean flow field based on SST and SAR provided by the present invention; Figure 2 Schematic diagram of the inversion method of direct back projection; Figure 3 Schematic diagram of the vector decomposition joint inversion method; Figure 4 is the radial velocity map calculated from SAR data; Figure 5 This is the two-dimensional flow field result diagram. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] See also Figure 1 The present invention provides a method for jointly constructing a two-dimensional ocean flow field based on SST and SAR. First, obtain the temporally and spatially matched SST data and SAR data; Then, the SST data is calculated using the MCC algorithm to obtain the change in ocean surface temperature and invert the flow angle of the ocean current field; the SAR data is calculated using the DCA algorithm to invert the radial velocity of the ocean current field; Finally, the flow direction angle and radial velocity of the ocean flow field are combined through the vector synthesis method to invert the two-dimensional flow field vector of the ocean flow field.

[0020] This method improves the spatial resolution and timeliness of ocean current field inversion.

[0021] Specifically, to obtain the time-space matching SST data and SAR data, the process is as follows: read two images and images The selected sea surface temperature data is in nc format. The SST data within the required latitude and longitude range is filtered out and saved as a CSV file for subsequent analysis.

[0022] The two images and images The sea surface temperature data are the sea surface temperature data of the same area at consecutive intervals.

[0023] SST data is acquired from the Himawari-8 remote sensing satellite, and SAR data is acquired from the Sentinel-1 remote sensing satellite. These SST and SAR data are spatially and temporally aligned to ensure the accuracy of the calculation results. The SST and SAR data must overlap in spatial coverage and be spaced no more than three hours apart.

[0024] The calculation process of SST data is as follows: the MCC algorithm is used to calculate the sea temperature gradient and variance within the window to determine whether the initial window meets the flow field velocity calculation requirements; if not, the window size is dynamically adjusted, and the sea temperature and flow velocity in the dynamic partition are recalculated in combination with the MCC algorithm.

[0025] The specific steps include: Step 11: For the two input images and images Perform mean removal processing and calculate the mean and , get a new image and new images : ; Step 12: Calculate the new image and new images The cross-correlation coefficient , used to measure the similarity between images; ; By maximizing the cross-correlation coefficient , get a preliminary estimate of the velocity field.

[0026] Step 13: Use the quadratic interpolation method to calculate the cross-correlation matrix Fitting is performed to achieve sub-pixel accuracy, thereby improving the resolution of the velocity field.

[0027] ; Step 14: Use the neighborhood average method to calculate the velocity vector , identify and correct outliers to improve the reliability and accuracy of the results. Specifically, for each velocity vector The calculation formula is as follows: ; in, Indicates the correlation coefficients selected from the spatial neighborhood when calculating the velocity vector of the target position. ≥R th The total number of all valid velocity vectors; where R th =0.65 is the preset cross-correlation coefficient threshold; are two images in the neighborhood and images The velocity vector inside.

[0028] Outliers include noise, incorrect estimates, etc.

[0029] Step 15: Use sliding window technology to select square windows of different sizes (such as 16×16 pixels, 32×32 pixels, or 64×64 pixels) for the image. and images Processing: For each position in the image, the velocity vector of the position is calculated using the windows of different sizes to obtain multi-scale velocity field data. For each window, steps 11 to 14 are repeated to extract the flow features at each scale and generate the velocity field vector. .

[0030] Step 16: Convert the velocity field vector The data is saved in NumPy format and provides multiple visualization methods, such as vector field maps, contour maps, and streamline maps.

[0031] In the present invention, the steps of extracting radial flow velocity from SAR data are as follows: Step 21: First, read and clean the SAR data. Then, collect the cleaned SAR data over multiple days and stitch and integrate the data for each orbit to form a continuous series of SAR spatiotemporal data. Verify data integrity to ensure there are no missing files. Integrate the multiple SAR images to create a unified coordinate system.

[0032] Step 22: extract the backscatter coefficient from the SAR spatiotemporal data and perform geometric correction.

[0033] The extracted backscatter coefficient is geometrically corrected to eliminate the influence of terrain and electromagnetic pointing errors.

[0034] In step 23, samples with excessively high wind speeds (greater than 10 m / s) are removed by combining the external wind field data from ECMWF. The uniform wind speed and direction distribution in the corresponding area of the SAR image is obtained through interpolation. The wind field data is provided by ECMWF.

[0035] The interpolation method is: ; in, is the interpolated velocity, is the weight, is the original wind speed at the measuring point.

[0036] The purpose of interpolating wind speed and direction is to improve the accuracy of flow velocity estimation.

[0037] Step 24: Calculate the radial velocity using the following formula based on the geometric information measured by SAR and the sea surface fluctuations: ; in, is the radial flow velocity, is the SAR wavelength, is the angle of incidence, is the Doppler shift.

[0038] In step 25, the CDOP (C-band Doppler Shift) model is used to correct the calculated radial velocity results to improve the accuracy of the final velocity results. The core formula of the CDOP model is: ; in, is the corrected speed, is a correction value affected by the wind field and incident angle to ensure that the final flow velocity is more consistent with the actual situation.

[0039] The results of the inversion of SST data using the MCC method are used to determine the true flow direction angle, and the results of the inversion of SAR data using the DCA method are used to determine the true flow velocity. The true flow velocity is mapped to the true direction, thereby inverting the accurate two-dimensional ocean flow field. Figures 2 to 5 , the specific steps for constructing the ocean two-dimensional flow field are as follows: In step 31, the flow direction angle extracted from the SST data by the MCC algorithm and the radial velocity results extracted from the SAR data by the DCA algorithm are resampled to grids of the same size using Gaussian filtering and bicubic interpolation methods.

[0040] Step 32: Based on the physical relationship between the components of the ocean current, derive the relevant formulas and use formulas (1) and (2) to calculate the meridional velocity of the two-dimensional flow field using the direct back projection method. , latitudinal speed , use formula (3) to calculate the longitudinal velocity of the two-dimensional flow field of the vector decomposition joint method , latitudinal speed ; (1) (2) (3) In the above formula, is the total vector of the two-dimensional flow field, including the flow velocity (unit: m / s) and the flow direction angle (unit: °, with due north as 0° and increasing clockwise); is the meridional vector of the two-dimensional flow field, is the latitudinal vector of the two-dimensional flow field, is the radial velocity vector, is the flow angle, is the radar viewing angle.

[0041] In step 33, the difference in the modulus lengths of the two velocity vectors is calculated according to the following formula: (4) in, ; It is the two-dimensional flow field vector inverted by direct back-projection method. It is the two-dimensional flow field vector inverted by the vector decomposition joint method. and represent the longitudinal velocity and latitudinal velocity of the two-dimensional flow field obtained based on the direct backprojection method, and They represent the longitudinal velocity and zonal velocity of the two-dimensional flow field obtained based on the vector decomposition joint method.

[0042] Step 34, calculate the correction factor. Use the correction factor to measure the speed deviation correction factor is calculated as follows: (5) in, Express the average velocity difference across all calculation points: (6) Step 35, correct the velocity component. Use the correction factor Weight the second set of velocity vectors to obtain the optimized velocity vector: (7) Using correction factors The initial flow field is corrected to improve the accuracy of flow field components and the reliability of the overall flow field.

[0043] The calculation results of the flow angle and radial flow velocity in the present invention can be used in the fields of marine environment monitoring, climate change research, marine disaster warning, etc.

[0044] Figure 2 middle, is the total vector of the two-dimensional flow field, and denote the longitudinal velocity and latitudinal velocity of the two-dimensional flow field obtained by direct back-projection method, is the radial velocity vector, is the flow angle, is the radar viewing angle.

[0045] Figure 3 middle, is the total vector of the two-dimensional flow field, and denote the longitudinal velocity and zonal velocity of the two-dimensional flow field obtained by the vector decomposition joint method, is the radial velocity vector, is the flow angle, is the radar viewing angle.

[0046] Figure 5 In the figure, the background shows the two-dimensional flow field velocity constructed by joint inversion of SST and SAR data, and the arrows indicate the direction and magnitude of the two-dimensional flow field vector in the area.

[0047] The present invention was verified by actual test tide stations, and an inversion accuracy of an average correlation coefficient of 0.72 and an angular phase difference of 5.57° was obtained at the "Dasha Fishery Station".

[0048] The MCC algorithm in the present invention is the maximum cross-correlation algorithm, and the DCA algorithm is the Doppler centroid frequency shift algorithm. Figure 1 .

[0049] This paper provides a method for jointly constructing a two-dimensional ocean current field based on SST and SAR. This method combines sea surface temperature (SST) data from the Himawari-8 satellite and synthetic aperture radar (SAR) data from the Sentinel-1 satellite, using a joint inversion model to calculate the two-dimensional current field. This effective combination and complementary nature of SST and SAR not only overcomes the limitations of single remote sensing methods but also fully leverages the unique advantages of different types of remote sensing data, improving the spatial resolution and accuracy of the inversion results. The average correlation coefficient for inversion accuracy can reach 0.72, and the angular phase difference can reach 5.57°. This method effectively fills the gap in sub-mesoscale ocean current data.

[0050] Of course, the present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for jointly constructing a two-dimensional ocean flow field based on SST and SAR, characterized in that: First, obtain the temporally and spatially matched SST data and SAR data; Then, the SST data is calculated using the MCC algorithm to obtain the change in ocean surface temperature and invert the flow angle of the ocean current field; the SAR data is calculated using the DCA algorithm to invert the radial velocity of the ocean current field; Finally, the flow direction angle and radial velocity of the ocean flow field are combined through the vector synthesis method to invert the two-dimensional flow field vector of the ocean flow field.

2. The method for jointly constructing a two-dimensional ocean flow field based on SST and SAR according to claim 1, characterized in that: To obtain the time-space matching SST data and SAR data, the process is as follows: read two images and images The sea surface temperature data is filtered out from the selected sea surface temperature data to obtain the SST data within the required longitude and latitude range and save it.

3. The method for jointly constructing an ocean two-dimensional flow field based on SST and SAR according to claim 2, characterized in that: SST data are acquired via the Himawari-8 remote sensing satellite, and SAR data are acquired via the Sentinel-1 remote sensing satellite. The SST data and SAR data must meet the following requirements: their spatial coverage overlaps, and their time interval does not exceed 3 hours.

4. The method for jointly constructing an ocean two-dimensional flow field based on SST and SAR according to claim 1, characterized in that: The calculation process of SST data is as follows: the MCC algorithm is used to calculate the sea temperature gradient and variance within the window to determine whether the initial window meets the flow field velocity calculation requirements; if not, the window size is dynamically adjusted, and the sea temperature and flow velocity in the dynamic partition are recalculated in combination with the MCC algorithm.

5. The method for jointly constructing a two-dimensional ocean flow field based on SST and SAR according to claim 4, characterized in that: The calculation process of SST data includes the following steps: Step 11: For the two input images and images Perform mean removal processing and calculate the mean and , get a new image and new images , the formula is as follows: ； Step 12: Calculate the new image and new images The cross-correlation coefficient , used to measure the similarity between images, the formula is as follows: ； By maximizing the cross-correlation coefficient, a preliminary estimate of the velocity field is obtained; Step 13: Use the quadratic interpolation method to calculate the cross-correlation matrix The fitting formula is as follows: ； Step 14: Use the neighborhood average method to calculate the velocity vector , identify and correct outliers; for each velocity vector The calculation formula is as follows: ； Among them, N represents the correlation coefficient selected from its spatial neighborhood when calculating the velocity vector of the target position. ≥R th The total number of all valid velocity vectors; where R th =0.65 is the preset cross-correlation coefficient threshold; are two images in the neighborhood and images Velocity vector inside; Outliers include noise and misestimation; Step 15: Use sliding window technology to use square windows of different sizes to image and images Processing; for each position in the image, using the windows of different sizes to respectively calculate the velocity vector of the position to obtain multi-scale velocity field data; Repeat steps 11 to 14 for each window to extract flow features at each scale and generate velocity field vectors. ; Step 16, save the velocity field vector .

6. The method for jointly constructing a two-dimensional ocean flow field based on SST and SAR according to claim 1, characterized in that: The steps for extracting radial velocity from SAR data are as follows: Step 21, reading and cleaning the SAR data; collecting the cleaned SAR data for multiple days, and splicing and integrating the data of each track to form a series of continuous SAR spatiotemporal data; Step 22, extracting backscatter coefficients from SAR spatiotemporal data and performing geometric correction; Step 23: Combine the external wind field ECMWF data, remove samples with wind speeds greater than 10 m / s, and obtain the uniform wind speed and wind direction distribution of the corresponding area of the SAR image through an interpolation method. The interpolation method is: ； in, is the interpolated velocity, is the weight, is the original wind speed at the measurement point; Step 24: Calculate the radial velocity using the following formula based on the geometric information measured by SAR and the sea surface fluctuations: ； in, is the radial flow velocity, is the SAR wavelength, is the angle of incidence, is the Doppler shift; Step 25: Use the CDOP model to correct the calculation results of the radial flow velocity. The core formula of the CDOP model is: ； in, is the corrected speed, is the correction value affected by wind field and incident angle.

7. The method for jointly constructing a two-dimensional ocean flow field based on SST and SAR according to claim 1, characterized in that: The results obtained by using the MCC method to invert the SST data are determined as the true flow direction angle, and the results obtained by using the DCA method to invert the SAR data are determined as the true flow velocity. The true flow velocity is mapped to the true direction, thereby inverting the accurate two-dimensional ocean flow field.

8. The method for jointly constructing a two-dimensional ocean flow field based on SST and SAR according to claim 7, characterized in that: The steps for constructing the two-dimensional ocean flow field are as follows: Step 31, the flow direction angle extracted from the SST data by the MCC algorithm and the radial velocity results extracted from the SAR data by the DCA algorithm are resampled to the same size grid through Gaussian filtering and bicubic interpolation method; Step 32: Based on the physical relationship between the components of the ocean current, derive the relevant formulas, use formulas (1) and (2) to calculate the meridional velocity and zonal velocity of the two-dimensional flow field of the direct back projection method, and use formula (3) to calculate the meridional velocity and zonal velocity of the two-dimensional flow field of the vector decomposition joint method; （1） （2） （3） In the above formula, It is the total vector of the two-dimensional flow field, including the flow velocity and flow angle. The unit of flow velocity is m / s, and the unit of flow angle is degrees. The flow angle is 0 degrees from due north and increases clockwise. is the meridional vector of the two-dimensional flow field, is the latitudinal vector of the two-dimensional flow field, is the radial velocity vector, is the flow angle, is the radar viewing angle; In step 33, the difference in the modulus lengths of the two velocity vectors is calculated according to the following formula: （4） in, ; It is the two-dimensional flow field vector inverted by direct back-projection method. It is the two-dimensional flow field vector inverted by the vector decomposition joint method. and represent the longitudinal velocity and latitudinal velocity of the two-dimensional flow field obtained based on the direct backprojection method, and They represent the meridional velocity and zonal velocity of the two-dimensional flow field obtained based on the vector decomposition joint method; Step 34, calculating the correction factor; Use correction factor to measure speed deviation correction factor is calculated as follows: （5） in, Express the average velocity difference of all calculation points: （6） Step 35, correcting the velocity component; Using correction factors Weight the second set of velocity vectors to obtain the optimized velocity vector: （7）。

Images