Internal solitary wave detection method based on sentry No.3 satellite altimeter data

Abstract

The invention relates to the field of ocean internal solitary wave automatic detection, and discloses an internal solitary wave detection method based on sentinel No.3 satellite altimeter data, which comprises the following steps: acquiring sentinel No.3 satellite altimeter data of a sea area to be detected, and preprocessing the data; sequentially extracting a posterior scattering coefficient of each coordinate point of the Ku wave band and the C wave band, and response characteristic parameters SSHA and SWH; sequentially calculating the difference mean square slopes of the posterior scattering coefficients of the Ku wave band and the C wave band; the response characteristic parameters are smoothed twice, and variables # imgabs0 # and # imgabs1 # are obtained; and respectively calculating a differential mean square slope and a local standard deviation of variables # imgabs2 # and # imgabs3 #, and extracting data fluctuation characteristics by setting a threshold value so as to realize internal solitary wave detection. The method disclosed by the invention has higher accuracy and lower false detection rate, and still has a good detection effect under the condition that the propagation direction of the inner solitary wave is greatly different from the running direction of the satellite.

Description

Technical Field

[0001] The present invention relates to the field of automatic detection of ocean internal solitary waves, and in particular to an internal solitary wave detection method based on altimeter data of a Sentinel-3 satellite. Background Art

[0002] During the propagation of ocean internal solitary waves, they carry huge energy and induce sudden strong currents and strong isopycnic surface fluctuations, with a maximum amplitude of up to 240m, which seriously threatens the navigation safety of submarines underwater. The maximum flow velocity caused by internal solitary waves exceeds 2m / s, which produces a large impact load on offshore oil platforms and engineering operations, and even causes them to move significantly, which has a huge impact on construction and installation, as well as safety or operational performance during operations. Therefore, the study of internal solitary waves is of great significance to marine science, national defense construction, marine engineering, etc.

[0003] When internal solitary waves propagate underwater, they can not only cause the convergence and divergence of microscale waves on the sea surface and be imaged in optical and synthetic aperture radar (SAR) remote sensing images, but also cause changes in the height fluctuation of the sea surface. The development of radar altimeters has provided a new direction for the detection of the vertical dimension of internal solitary waves. The sea surface height fluctuation modulated by internal solitary waves is manifested in the altimeter data as changes in the sea surface height anomaly (SSHA) and the significant wave height (SWH).

[0004] Early altimeters were pulse-limited traditional altimeters with an along-track resolution of usually 2-3km and a sampling frequency of 1Hz. The footprint was too large, making it difficult to obtain information on sea surface height changes caused by internal solitary waves. The Synthetic Aperture Radar Altimeter (SRAL) carried by the Sentinel-3 satellite has an along-track resolution of 300m and provides 20Hz data products, which can continuously record the characteristics of sea surface height changes caused by internal solitary waves. However, the current methods are mainly based on the mean square slope of the backscatter coefficient and the peak power of the waveform for detection, and the results are not ideal. At the same time, when the propagation direction of the internal solitary wave is significantly different from the direction of satellite operation, the sea surface echo signal detected by the SRAL is interfered by multiple factors, the received internal solitary wave signal is weak, and the internal solitary wave is even more difficult to detect. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an internal solitary wave detection method based on the altimeter data of the Sentinel-3 satellite, so as to achieve the purpose of improving the detection accuracy.

[0006] To achieve the above object, the technical solution of the present invention is as follows: A method for detecting internal solitary waves based on Sentinel-3 satellite altimeter data comprises the following steps: Step 1, obtain the altimeter data of the Sentinel-3 satellite in the sea area to be measured and pre-process the data; Step 2: Based on the preprocessed data, extract the backscattering coefficient of each coordinate point in the Ku band in turn. And the backscattering coefficient of each coordinate point in the C band , and the response characteristic parameters SSHA and SWH, where SSHA and SWH are the sea surface height anomaly and significant wave height, respectively; Step 3: Calculate the differential mean square slope of the backscattering coefficients of the Ku band and C band corresponding to each coordinate point in turn. ; Smoothing the response characteristic parameters SSHA and SWH twice, we get two new variables and ; Step 4: Calculate the differential mean square slope of each coordinate point And two new variables and The local standard deviation of the data is used to extract the data fluctuation characteristics by setting the threshold to realize the detection of internal solitary waves.

[0007] In the above scheme, in step 1, the data is preprocessed by removing the data within 3 km of the land and the land boundary, and converting the backscattering coefficient and The value range of is set to 8~18, and the data range of the response characteristic parameter SSHA is set to , and remove water vapor content greater than 60Kg / m 2 and liquid water concentration greater than 0.1Kg / m 2 The abnormal backscattering coefficient and SSHA value of .

[0008] In the above scheme, in step 3, the coordinate point is calculated The corresponding differential mean square slope of the backscattering coefficients in the Ku band and C band The calculation is as follows: ; in, is the Ku band coordinate point The mean square slope of is the C band coordinate point The mean square slope of is the Ku band coordinate point The effective reflectivity coefficient, is the Ku band coordinate point The backscattering coefficient of is the C band coordinate point The effective reflectivity coefficient, is the C band coordinate point The backscattering coefficient of is the bias parameter, is the C band coordinate point The calibration value of the backscattering coefficient.

[0009] In the above scheme, in step 3, the specific method of performing two smoothing processes on the response characteristic parameters SSHA and SWH is as follows: the moving average method is used to perform two smoothing processes, and the smoothing length is set to 1.8 km for the first smoothing process; the smoothing length is set to 30 km for the second smoothing process; then the difference between the first smoothing result and the second smoothing result is calculated to obtain two new variables and , is the abnormal change of sea surface height after smoothing, is the effective wave height change after smoothing.

[0010] In the above scheme, in step 4, the coordinate point is calculated The difference mean square slope And two new variables and The local standard deviation The formula is as follows: ; Where W is the window length, Representative coordinate points, Represents the window i The difference mean square slope of the coordinate points ,variable or ; is the differential mean square slope within the window ,variable or The local average of the data is calculated as follows: ; The difference mean square slope is obtained by calculating ,variable and The local mean , and , and the difference mean square slope ,variable and The local standard deviation , and .

[0011] In the above scheme, the threshold calculation formula is as follows: ; Where K is the coefficient; when the local standard deviation The coefficient of dispersion When K=2, When K=1, the coefficient of dispersion The calculation formula is as follows: ; According to the coefficient K, the local standard deviation is calculated respectively , and The corresponding threshold , and , respectively, the local standard deviation , and The data that are greater than the corresponding thresholds are extracted, and finally the detection results of the three items of data are intersected and output as the location where the detected internal solitary wave occurs.

[0012] Through the above technical solution, the internal solitary wave detection method based on Sentinel-3 satellite altimeter data provided by the present invention has the following beneficial effects: 1. Compared with the existing small amount of research on detecting internal solitary waves based on altimeter data, the present invention adopts the dual-band characteristics of the altimeter and is established based on the wave characteristics of the data. It has a higher accuracy and a lower false detection rate in the detection of internal solitary waves, and can detect internal solitary waves of various types and different scales, which has a higher research value; 2. Traditional methods usually rely only on mean square slope or waveform peak power to achieve automatic detection of internal solitary waves, while the present invention combines the SRAL data with the The internal solitary wave feature detection is realized by using the comprehensive response characteristics such as SSHA and SWH. The detection threshold is set to be adaptively adjusted according to different situations, which improves the accuracy of internal solitary wave detection. 3. Compared with the traditional method, the present invention not only eliminates the influence of land and land boundary reflection signals in preprocessing, but also eliminates the interference of oil spills, rainfall and other phenomena, and realizes the processing of outliers in SRAL data that do not belong to the characteristics of internal solitary waves; 4. Compared with the traditional method of directly using mean square slope to detect internal solitary waves, based on C band The algorithm is easy to cause missed detection, and the Ku-band The invention extracts characteristic parameters of Ku band and C band from different bands. , can more fully utilize all-round information and improve the accuracy of detection algorithms; 5. The present invention selects the moving average method to first smooth SSHA and SWH to filter out high-frequency disturbances in the data; then another smoothing process is performed on the original SSHA and SWH to eliminate the influence of medium and large-scale ocean phenomena; finally, the difference between the first smoothing result and the second smoothing result is calculated to obtain two new variables and This variable eliminates the impact of medium- and large-scale ocean phenomena on the data, filters out high-frequency disturbances in the data that are not internal solitary waves, and its "resolution" is suitable for the scale of internal solitary waves; 6. The present invention makes comprehensive use of the internal solitary wave , SSHA and SWH change characteristics, calculated the differential mean square slope and local standard deviation to highlight the internal solitary wave characteristics in the data, and used the adaptive threshold to extract the internal solitary wave characteristics in the data to complete the detection. Compared with the previous methods, this method has higher accuracy, lower false detection rate, and is applicable to more sea areas. When the propagation direction of the internal solitary wave is significantly different from the satellite operation direction, this algorithm still has a good detection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0014] Figure 1 A schematic flow chart of an internal solitary wave detection method based on Sentinel-3 satellite altimeter data disclosed in an embodiment of the present invention; Figure 2 This is the OLCI optical remote sensing image of the Sulu Sea taken by Sentilne-3 on April 6, 2020; Figure 3 For Figure 2 The OLCI optical remote sensing image is matched with the Sentilne-3 SRAL data of the Sulu Sea. (a) is the backscattering coefficient (a) is the variation curve of the sea surface height anomaly SSHA, and (c) is the variation curve of the significant wave height SWH; Figure 4 The internal solitary wave detection results of Sentilne-3 SRAL data in the Sulu Sea on April 6, 2020; (a) is the differential mean square slope variable The local standard deviation curve of (b) is the sea surface height anomaly variable The local standard deviation curve of (c) is the effective wave height variable The local standard deviation change curve, (d) is the final detection result of this method; Figure 5 The internal solitary wave detection results of Sentilne-3 SRAL data in the Sulawesi Sea on September 15, 2019; Figure 6 for Figure 5 Synchronous OLCI images of the test results; Figure 7 The internal solitary wave detection results of Sentilne-3 SRAL data in the South China Sea on April 25, 2023; Figure 8 for Figure 7 Quasi-synchronous field observation data of test results; Fig. 9 The results of internal solitary wave detection in Sentilne-3 SRAL data on November 30, 2023 in northeastern Brazil; Fig.10 for Fig. 9 Quasi-synchronous SWOT data of the test results. DETAILED DESCRIPTION

[0015] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0016] The present invention provides an internal solitary wave detection method based on the altimeter data of the Sentinel-3 satellite. Figure 1 As shown, the following steps are included: Step 1: Obtain the Sentinel-3 satellite altimeter data in the Sulu Sea and preprocess the data.

[0017] The satellite altimeters used in the embodiment construct a data set of Sentinel-3A data with a total of 581 passes in 9 relative orbits from 2017 to 2023 and Sentinel-3B data with a total of 376 passes in 7 relative orbits from 2019 to 2023.

[0018] Get the OLCI optical remote sensing images of the Sulu Sea collected by the Ocean and Land Color Instrument (OLCI) on the Sentinel-3 satellite, such as Figure 2 Taking advantage of the synchronous observation characteristics of OLCI and SRAL, 129 altimeter data containing internal solitary waves that have been verified by OLCI optical remote sensing images are extracted from the data set, and 178 internal solitary wave samples are extracted.

[0019] The optical remote sensing image data in the data set is processed by geometric correction, optical enhancement, etc., and then the internal solitary wave samples are extracted based on the processed optical remote sensing image. This step mainly processes the OLCI optical remote sensing image data that is synchronized with the SRAL data. Because the OLCI optical remote sensing image is easily disturbed by the external environment such as clouds and fog, the performance characteristics of the internal solitary wave in the remote sensing image are often covered or unclear. Through optical enhancement processing, the best internal solitary wave visibility information can be obtained, which is then used for the subsequent verification of the detection method of the present invention.

[0020] Step 2: Based on the preprocessed Sentinel-3 SRAL data, extract the backscattering coefficient of each coordinate point in the Ku band in turn. And the backscattering coefficient of each coordinate point in the C band , and the response characteristic parameters SSHA and SWH, where SSHA and SWH are the sea surface height anomaly and significant wave height, respectively. Figure 3 shown.

[0021] In SAR mode, the footprint of SRAL is a rectangle with a width of 300m and a length of 2-3km. When the altimeter detects the sea surface at the land boundary, although the sub-satellite point of the altimeter detects the sea surface, the altimeter footprint will still receive the signal reflected from the land. At this time, the backscattering coefficient The data on land and land boundaries need to be removed because the data will be different from those in the open sea, which will interfere with the detection method. Combined with the characteristics of SRAL footprints, the data within 3km of land and land boundaries are removed.

[0022] In addition to land, some special phenomena can also cause the backscattering coefficient Abnormal behavior, such as oil spills in the ocean, can cause the sea surface roughness to decrease, resulting in Abnormally large. It is necessary to further eliminate abnormally large and small In order to exclude the interference of oil spills and other phenomena on the algorithm. The SSHA changes caused by internal solitary waves are usually at the decimeter level, while the SSHA changes caused by storm surges and other phenomena are usually at the meter level. Therefore, in order to exclude the interference of special phenomena that are not internal solitary waves on the algorithm, the following will be effective and the SSHA range is taken as: ; .

[0023] After excluding the influence of land and special phenomena on the data, rainfall is also an important factor that needs to be considered in the detection method. The radar altimeter signal will be absorbed and scattered by raindrops, resulting in signal attenuation and abnormal changes in the altimeter data. In addition, the Ku-band data is more affected by rainfall than the C-band data. Sentinel-3 SRAL data provides two data, "water vapor" and "liquid water" content. The following sets the range of water vapor content and liquid water concentration to: ; ; In the formula, is the water vapor content, ; is the integrated columnar liquid water content, Therefore, the water vapor content is greater than 60Kg / m 2 and liquid water concentration greater than 0.1Kg / m 2 The abnormal backscattering coefficient and SSHA value of .

[0024] The above steps complete the preprocessing of SRAL data.

[0025] Step 3: Calculate the differential mean square slope of the backscattering coefficients of the Ku band and C band corresponding to each coordinate point in turn. ; and the response characteristic parameters SSHA and SWH are smoothed twice to obtain two new variables and .

[0026] The SRAL of the Sentinel-3 satellite is a dual-frequency radar altimeter, so the data products provide both Ku-band and C-band. The sensitivity of Ku-band and C-band pulse signals to changes in sea surface roughness is significantly different. The sensitivity of C-band pulse signals is lower, and the changes in sea surface roughness caused by small and medium-scale solitary waves are smaller, resulting in its sensitivity in the C-band. It is difficult to show obvious fluctuations in the Ku-band pulse signal. The Ku-band pulse signal is more sensitive to changes in sea surface roughness, so even small-scale internal solitary waves will have certain fluctuations in the data. However, this also makes the data susceptible to rainfall and other small and medium-scale ocean phenomena. Therefore, based on the C-band The algorithm is easy to cause missed detection, and the Ku-band Establishing an algorithm is prone to cause false detection. To address the above problem, a detection method is constructed by calculating the differential mean square slope.

[0027] Calculate coordinate points The corresponding differential mean square slope of the backscattering coefficients in the Ku band and C band The calculation is as follows: ; in, is the Ku band coordinate point The mean square slope of is the C band coordinate point The mean square slope of is the Ku band coordinate point The effective reflectivity coefficient, is the Ku band coordinate point The backscattering coefficient of is the C band coordinate point The effective reflectivity coefficient, is the C band coordinate point The backscattering coefficient of is the C band coordinate point The calibration value of the backscattering coefficient is based on the research results of Frew et al. , and Take 0.427, 0.617 and 3.61. It is worth noting that the above parameters are calculated for the TOPEX altimeter, so a bias parameter needs to be introduced To compensate for the C-band difference between TOPEX and Sentinel-3A / B radar altimeters In this embodiment, .

[0028] For Sentinel-3A / B dual-frequency altimeter, Ku-band , C-band , the corresponding cut-off wavelengths are 6.3cm and 16.5cm respectively. By differentiating the estimated values ​​of the mean square slope of the two frequency bands, the mean square slope contribution of small-scale waves between 6.3cm and 16.5cm can be extracted, and the modulation effect of short-period solitary waves on the surface wave spectrum is strongest in this wavelength range. The footprint size of the Ku-band and C-band frequencies of the Sentinel-3A / B altimeter is the same, which indicates that the pulse sampling area of ​​the radar in the two frequency bands is the same. Therefore, by calculating the differential mean square slope, the backscatter contribution from the longer waves can be eliminated, highlighting the The internal solitary wave characteristics in .

[0029] For SSHA and SWH, the "ssha_20_ku" and "swh_20_ku" products in the Sentinel-3 SRAL data L2 level are selected. These two data provide the actual measurement values ​​of the vertical displacement of the sea surface and the effective wave height of solitary waves in a short period by the satellite. The raw data of SSHA and SWH will be affected by small-scale sea surface disturbances, and need to be filtered to filter out high-frequency disturbances in the time series and retain useful low-frequency trends.

[0030] In this embodiment, the specific method of performing two smoothing processes on the response characteristic parameters SSHA and SWH is as follows: the moving average method is used to perform two smoothing processes. The area size of the sea surface height field modulated by the internal solitary wave is usually greater than 2 km. Therefore, the smoothing length is set to 1.8 km in the first smoothing process to filter out high-frequency disturbances in the data; medium- and large-scale ocean phenomena can also cause abnormal changes in SSHA and SWH. In order to eliminate the influence of medium- and large-scale ocean phenomena, the smoothing length is set to 30 km in the second smoothing process; then the difference between the first smoothing result and the second smoothing result is calculated to obtain two new variables and , is the abnormal change of sea surface height after smoothing, is the effective wave height change after smoothing. This variable eliminates the influence of medium and large scale ocean phenomena on the data, filters out high-frequency disturbances in the data that do not belong to internal solitary waves, and its "resolution" is suitable for the scale of internal solitary waves.

[0031] Step 4: Calculate the differential mean square slope of each coordinate point And two new variables and The local standard deviation of the data is used to extract the data fluctuation characteristics by setting the threshold to realize the detection of internal solitary waves.

[0032] Calculate the difference mean square slope ,variable and To calculate the local standard deviation, we first need to determine the size of the window W, which is determined by the crest line width of the internal solitary wave. The local standard deviation is calculated to highlight the data fluctuations caused by the internal solitary wave. Therefore, the length of the window should be similar to the crest line width of the internal solitary wave, and it should not be too large or too small. It is most appropriate to choose the medium-scale crest line width of the internal solitary wave as the length of the window. The window length selected by this algorithm is 3.9km, that is, W=13. After determining the size of the window W, the local average value of the data is calculated using the following formula: ; in, Represents the window i The difference mean square slope of the coordinate points ,variable or ; The local standard deviation of the data is further calculated by the local mean , the formula is as follows: ; Where W is the window length, Representative Coordinate points.

[0033] The difference mean square slope is obtained by calculating ,variable and The local mean , and , and the difference mean square slope ,variable and The local standard deviation , and .

[0034] At this time, the isolated area , and Compared with the background sea surface, they all show more significant fluctuation characteristics. By setting a threshold to extract the data fluctuation characteristics, internal solitary wave detection can be achieved. This method refers to the distribution law of data in the normal distribution. In order to extract abnormal fluctuations in the data, the threshold of the algorithm detection is obtained by calculating the mean standard deviation of the data. The threshold calculation formula is as follows: ; Among them, K is the coefficient; , and middle, The change characteristics of are the most obvious. In order to further improve the accuracy of the threshold X, this method calculates The coefficient of dispersion To determine the coefficient K. The coefficient of dispersion is calculated by the following formula: ; Coefficient of dispersion It reflects the volatility of the data. When , the data fluctuation is small, indicating that there may be no internal solitary waves or small-scale internal solitary waves in the detection range. In order to extract only the data fluctuation caused by small-scale internal solitary waves, K=2. When , it indicates that the data is more discrete. In order to extract the abnormal fluctuation characteristics, K=1.

[0035] According to the coefficient K, the local standard deviation is calculated respectively , and The corresponding threshold , and , respectively, the local standard deviation , and The data greater than the corresponding thresholds are extracted, and finally the detection results of the three data are intersected and output as the location of the detected internal solitary wave. The internal solitary wave detection results of the Sentilne-3 SRAL data in the Sulu Sea on April 6, 2020 are as follows: Figure 4 shown.

[0036] Finally, using the Sentinel-3 altimeter data, internal solitary waves were detected in the Sulawesi Sea on September 15, 2019, the South China Sea on April 25, 2023, and the northeastern Brazilian waters on November 30, 2023. The detection results are as follows: Figure 5 , Figure 7 and Fig. 9 As shown, the method of the present invention is verified using the synchronized optical OLCI data of step 1, the quasi-synchronous field observation data, and the SWOT data of the three-dimensional imaging radar altimeter. The verification results are as follows: Figure 6 , Figure 8 and Fig.10 As shown. Figures 5 to 10 It can be seen that the method of the present invention has high accuracy and universality in detecting internal solitary waves, and has a low false detection rate.

[0037] The test results were verified using the OLCI optical remote sensing images of Sentinel-3, and the data coverage areas include the Sulu Sea, the Sulawesi Sea, the Lombok Strait and the northeastern waters of Brazil. The method of the present invention detected 164 out of 178 positive samples (internal solitary wave samples), missed 14, and misdetected 3, with a detection accuracy of 92.2%. This shows that the detection method has a high accuracy rate and a low misdetection rate, can effectively detect marine internal solitary waves, and is suitable for different sea areas. In addition, for the Sulu Sea, the Sulawesi Sea, the Lombok Strait and the northeastern waters of Brazil, the detection results of the existing mean square slope method based on the backscattering coefficient are that 113 samples were detected, 65 were missed, 37 were misdetected, and the detection accuracy was 63.4%. The detection results of the waveform peak power method are that 152 samples were detected, 26 were missed, 23 were misdetected, and the accuracy was 85.4%. It can be seen that the method of the present invention has higher detection accuracy.

[0038] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting internal solitary waves based on Sentinel-3 satellite altimeter data, characterized in that: The steps include: Step 1, obtain the altimeter data of the Sentinel-3 satellite in the sea area to be measured and pre-process the data; Step 2: Based on the preprocessed data, extract the backscattering coefficient of each coordinate point in the Ku band in turn. And the backscattering coefficient of each coordinate point in the C band , and the response characteristic parameters SSHA and SWH, where SSHA and SWH are the sea surface height anomaly and significant wave height, respectively; Step 3: Calculate the differential mean square slope of the backscattering coefficients of the Ku band and C band corresponding to each coordinate point in turn. ; Smoothing the response characteristic parameters SSHA and SWH twice, we get two new variables and ; Step 4: Calculate the differential mean square slope of each coordinate point And two new variables and The local standard deviation of the data is used to extract the data fluctuation characteristics by setting the threshold to realize the detection of internal solitary waves.

2. The internal solitary wave detection method based on Sentinel-3 satellite altimeter data according to claim 1, characterized in that: In step 1, the data is preprocessed by removing the data within 3 km of the land and land boundary, and converting the backscattering coefficient and The value range of is set to 8~18, and the data range of the response characteristic parameter SSHA is set to , and remove water vapor content greater than 60Kg / m 2 and liquid water concentration greater than 0.1Kg / m 2 The abnormal backscattering coefficient and SSHA value of .

3. The internal solitary wave detection method based on Sentinel-3 satellite altimeter data according to claim 1, characterized in that: In step 3, calculate the coordinate points The corresponding differential mean square slope of the backscattering coefficients in the Ku band and C band The calculation is as follows: ； in, is the Ku band coordinate point The mean square slope of is the C band coordinate point The mean square slope of is the Ku band coordinate point The effective reflectivity coefficient, is the Ku band coordinate point The backscattering coefficient of is the C band coordinate point The effective reflectivity coefficient, is the C band coordinate point The backscattering coefficient of is the bias parameter, is the C band coordinate point The calibration value of the backscattering coefficient.

4. The internal solitary wave detection method based on Sentinel-3 satellite altimeter data according to claim 1, characterized in that: In step 3, the specific method of performing two smoothing processes on the response characteristic parameters SSHA and SWH is as follows: the moving average method is used for two smoothing processes. The smoothing length is set to 1.8 km for the first smoothing process and 30 km for the second smoothing process. Then, the difference between the first smoothing result and the second smoothing result is calculated to obtain two new variables: and , is the abnormal change of sea surface height after smoothing, is the effective wave height change after smoothing.

5. The internal solitary wave detection method based on Sentinel-3 satellite altimeter data according to claim 1, characterized in that: In step 4, calculate the coordinate points The difference mean square slope And two new variables and The local standard deviation The formula is as follows: ； Where W is the window length, Representative coordinate points, Represents the differential mean square slope of the i-th coordinate point in the window ,variable or ; is the differential mean square slope within the window ,variable or The local average of the data is calculated as follows: ； The difference mean square slope is obtained by calculating ,variable and The local mean , and , and the difference mean square slope ,variable and The local standard deviation , and .

6. The internal solitary wave detection method based on Sentinel-3 satellite altimeter data according to claim 5, characterized in that: The threshold calculation formula is as follows: ； Where K is the coefficient; when the local standard deviation The coefficient of dispersion When K=2, When K=1, the coefficient of dispersion The calculation formula is as follows: ； According to the coefficient K, the local standard deviation is calculated respectively , and The corresponding threshold , and , respectively, the local standard deviation , and The data that are greater than the corresponding thresholds are extracted, and finally the detection results of the three items of data are intersected and output as the location where the detected internal solitary wave occurs.

Images