Wave wavelength dispersive spectrometer nadir point wind speed inversion method

Abstract

The invention discloses a nadir point wind speed inversion method of a sea wave spectrometer. The nadir point wind speed inversion method specifically comprises the steps of 1, reading parameters used for nadir point sea surface wind speed inversion in SWIM L2-level product data; 2, the read data are screened, and effective data are reserved; step 3, enabling the nadir point standardized backscattering coefficient to correspond to the box where the nadir point standardized backscattering coefficient is located based on the nadir point box index; 4, calculating an average sea surface mean square slope according to a linear fitting coefficient of the NRCS profile; 5, calculating a spectrum energy level according to the one-dimensional sea wave spectrum; 6, normalizing the nadir point standardized backscattering coefficient, the average sea surface mean square slope of the box where the nadir point standardized backscattering coefficient is located and the spectral energy level of the box where the nadir point standardized backscattering coefficient is located; and calculating the sea surface wind speed of the nadir point based on the normalized data. The method can further improve the calculation precision.

Description

Technical Field

[0001] The invention belongs to the technical field of marine engineering, and in particular relates to a method for inverting nadir wind speed of an ocean wave spectrometer. Background Art

[0002] Ocean wind field information is one of the important research contents in ocean dynamics and meteorology, and is of great significance to global climate change prediction, marine disaster warning and marine resource development. Accurate measurement of wind speed is crucial for marine research, ship navigation, weather forecast and military activities. Traditional methods of obtaining sea surface wind fields include observation station observation, ship observation and buoy observation. Although these observation methods have high accuracy, they have disadvantages such as low spatial coverage, uneven distribution, long measurement cycle and poor repeatability. Ocean remote sensing satellite observation can obtain sea surface wind field information of large spatial scale in a short time, provide global sea surface wind field data and realize real-time monitoring of sea surface wind field, which makes up for the shortcomings of traditional methods and becomes one of the main means of obtaining global sea surface wind field.

[0003] The wave spectrometer SWIM carried by the ocean satellite CFOSAT (China France Oceanography Satellite) is the world's first instrument specifically used to measure two-dimensional ocean wave spectra. The wave spectrometer SWIM (Surface Waves Investigation and Monitoring) is a multi-beam true aperture radar working in the Ku band, with 6 detection beams with different incident angles, and the incident angles of the beam center are 0°, 2°, 4°, 6°, 8°, and 10° respectively. Except for the 0° beam, the other beams adopt a rotating scanning working mode, which can obtain the sea surface backscattering coefficients in different directions under different incident angles. The sea surface backscattering coefficient is a key parameter for inverting the sea surface wind field. Since the mean square slope of the sea surface short-scale waves is highly correlated with the sea surface wind speed, there is a nonlinear inverse relationship between the backscattering coefficient and the sea surface wind speed. However, there is an error between the sea surface wind speed calculated based on a single backscattering coefficient and the actual wind speed. The reason for the error is the fluctuation of the medium and long scales of the sea surface. These medium- and long-scale sea surface fluctuations are not affected by the local sea surface, but they will cause the short-scale sea surface waves to tilt to a certain extent, thereby causing changes in the backscattering coefficient observed by satellite.

[0004] Therefore, adding the sea surface effective wave height and the backscattering coefficient to collaboratively invert the sea surface wind speed is currently a popular altimeter wind speed inversion method. However, relying solely on the effective wave height cannot accurately characterize the fluctuations of the sea surface at different scales. The effective wave height is only a limited substitute parameter for considering the influence of waves in the altimeter wind speed inversion. Summary of the invention

[0005] Purpose of the invention: In order to solve the problems existing in the above-mentioned prior art, the present invention provides a method for inverting the wind speed at the nadir point of an ocean wave spectrometer.

[0006] Technical solution: The present invention provides a method for inverting the wind speed at the nadir point of an ocean wave spectrometer, which specifically comprises the following steps: Step 1: Read the parameters used for nadir sea surface wind speed inversion in SWIM L2 product data; including: surface type identifier, nadir rainfall identifier, nadir box index, nadir normalized backscatter coefficient, nadir normalized backscatter coefficient identifier, linear fitting coefficient of NRCS profile, wave spectrum number, one-dimensional wave spectrum, land surface proportion within the observation range, and sea ice surface proportion within the observation range; Step 2: Filter the read data and retain valid data; Step 3: Based on the nadir box index, the nadir point normalized backscatter coefficient corresponds to the box where it is located; Step 4: Calculate the mean sea surface mean square slope based on the linear fitting coefficient of the NRCS profile ; Step 5: Calculate the spectrum energy level based on the one-dimensional wave spectrum ; Step 6: Normalize the nadir point standardized backscatter coefficient, the average sea surface mean square slope of the box where the nadir point standardized backscatter coefficient is located, and the spectral energy level of the box where the nadir point standardized backscatter coefficient is located; calculate the nadir point sea surface wind speed based on the normalized data.

[0007] Furthermore, the step 2 is specifically as follows: data quality control is performed according to the surface type identifier, the nadir rainfall identifier, the nadir normalized backscatter coefficient and the wave spectrum product quality identifier, and non-ocean surface data and data with quality lower than a preset value are removed. The wave spectrum product quality identifier includes the proportion of land surface within the observation range and the proportion of sea ice surface within the observation range.

[0008] Furthermore, the step 4 is calculated by the following method: : Step 4.1: Calculate the mean square slope of the sea surface along the radar observation direction : ; Where B represents the slope of the linear fitting coefficient of the NRCS profile, and e represents the exponent; Step 4.2: Calculate the average sea surface mean square slope of all radar observation directions on one side of the nadir track for each box, and use the average value as the average sea surface mean square slope on one side of the box. Then calculate the average of the average sea surface mean square slopes of the box on both sides of the track to obtain the average sea surface mean square slope corresponding to the box. .

[0009] Furthermore, step 5 adopts the following method: first convert the slope spectrum in the one-dimensional wave spectrum into a height spectrum, then convert the height spectrum into a frequency spectrum F(f) according to the deep water wave dispersion relationship, and calculate the spectral energy level E of the box on one side of the nadir orbit based on the frequency spectrum according to the following formula: ; Among them, j represents the jth spectral density value of the one-dimensional wave spectrum, f represents the frequency, It represents the average value of the negative fourth power of the 15th to 32nd spectral density detection frequency of the spectrometer's one-dimensional wave spectrum. It represents the average value of the 15th to 32nd spectral density of the one-dimensional wave spectrum of the spectrometer. and The expression is as follows: ; ; Calculate the average of the spectral energy levels of the box on both sides of the nadir orbit .

[0010] Furthermore, the slope spectrum S(k) is converted into the height spectrum F(k) according to the following formula: ; Among them, k represents the wave number corresponding to the frequency; The wave number k is converted into frequency f according to the following formula: ; The height spectrum is converted to the frequency spectrum according to the following formula: ; in, Represents the acceleration due to gravity.

[0011] Furthermore, step 6 is normalized according to the following formula: ; in, represents the normalized backscatter coefficient of the nadir point after normalization. represents the initial nadir normalized backscatter coefficient value, represents the normalized mean square slope of the sea surface, represents the normalized spectral energy level.

[0012] Furthermore, step 6 calculates the sea surface wind speed at the nadir point according to the following formula: : ; in, is the intermediate variable matrix, and the expression is as follows: ; in, represents the weight matrix, represents the intermediate variable matrix, is the bias matrix, ; and The expression is as follows: ; ; in, represents the weight matrix, represents the bias matrix, represents the input matrix, , and The expression is as follows: ; ; ; in, represents the normalized backscatter coefficient of the nadir point after normalization. represents the normalized mean square slope of the sea surface, represents the normalized spectral energy level.

[0013] Beneficial effects: The present invention uses the wave spectrum and the mean square slope of the sea surface instead of the effective wave height to invert the nadir sea surface wind speed, which can further improve the calculation accuracy. The method of the present invention makes full use of the observation data of SWIM and increases the scope of application of SWIM products. The high-precision and high-resolution nadir wind speed calculated by the present invention can be used to correct and supplement the wind field information of the ocean satellite scatterometer, and can provide more accurate wind field data for studying climate change. In the future, new wind and wave satellites can use this method to calculate high-precision nadir sea surface wind speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of the present invention.

[0015] Figure 2This is a comparison diagram of the wind speed in October 2024 calculated according to the algorithm of the present invention and the ERA5 wind speed.

[0016] Figure 3 This is a comparison of the nadir wind speed of SWIM L2 v7.0 in October 2024 and the RA5 wind speed. DETAILED DESCRIPTION

[0017] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0018] like Figure 1 As shown, the process of a method for inverting the nadir wind speed of a wave spectrometer provided by the present invention is as follows: Step 1: Read the parameters for nadir sea surface wind speed inversion in the SWIM L2 product data. The read parameters are shown in Table 1 below: Table 1 Variable Name Specific instructions wf_surf_flag Surface type identifier: a value of 0 indicates that the surface is ocean, a value of 1 indicates that the surface is land, and a value of 2 indicates that the surface is sea ice nadir_rain_flag Nadir rainfall indicator: a value of 0 indicates no rainfall, a value of 1 indicates sigma0-b1oom, and a value of 2 indicates rainfall indices_boxes Nadir box index: the box index where the nadir point is located nadir_sigma0_native Nadir point normalized backscatter coefficient value nadir_sigma0_native_validity Nadir point normalized backscatter coefficient flag: a value of 0 indicates that the backscatter coefficient value is available, and a value other than 0 indicates that the backscatter coefficient value may have a large error coeff_sigma0_trend Linear fitting coefficients of NRCS profiles k_spectra Wave spectrum wave number pp_omni One-dimensional wave spectrum, including the wave spectra measured by the 6, 8, and 10 degree beams of the spectrometer Land_coverage The proportion of land surface within the observation range Sea_ice_coverage Sea ice surface percentage within the observation range In Table 1, coeff_sigma0_trend, k_spectra, pp_omni, Land_coverage, and Sea_ice_coverage represent the spectrometer ocean wave products, the geographic location unit is box, and the rest are ocean wave spectrum nadir products.

[0019] Step 2: Use the surface type identifier, rainfall identifier, nadir normalized backscatter coefficient and wave spectrum product quality identifier in the SWIM L2 product to control data quality, retain only valid data, and remove non-ocean surface data and data with poor quality (i.e., data below the preset value); the wave spectrum product quality identifiers include Land_coverage and Sea_ice_coverage.

[0020] Step 3: Use the nadir box index in the SWIM L2 product to match the nadir normalized backscatter coefficient to the box it is in.

[0021] Step 4: Calculate the mean sea surface mean square slope using the linear fitting coefficients of the Normalized Radar Cross Section (NRCS) profile read from the SWIM L2 data , the specific steps are as follows: Step 4-1: Read the linear fitting coefficient of the NRCS profile of SWIM L2 data, and calculate the mean square slope of the sea surface along the radar observation direction according to the following formula: : ; in: represents the slope of the linear fitting coefficient of the NRCS profile read from the SWIM L2 data, and e represents the exponent.

[0022] Step 4-2: Average the mean square slope of the sea surface in all radar observation directions on one side of the nadir orbit for each box (each box contains 12 different directions on one side of the nadir orbit, and due to the rotation characteristics of the spectrometer, the 12 directions are not fixed), and use the calculated average value as the average mean square slope of the sea surface on one side of the box. Then calculate the average of the mean square slopes of the average sea surface on both sides of the orbit of the box to obtain the average mean square slope of the sea surface corresponding to the box. .

[0023] Step 5: Use the one-dimensional wave spectrum observed with a 10-degree beam read from the SWIM L2 data to estimate its spectral energy level , the specific steps are as follows: Step 5-1: Convert the slope spectrum S(k) of the one-dimensional wave spectrum read from SWIM L2 data into the height spectrum F(k) ;

[0024] Where k represents the wave number corresponding to the frequency.

[0025] Step 5-2: According to the deep-water wave dispersion relation, convert the wavenumber spectrum into a frequency spectrum: ; ; Where f is the frequency, represents the acceleration due to gravity, and F(f) represents the frequency spectrum.

[0026] Step 5-3: Use the 15th to 32nd spectral density values ​​of the frequency spectrum F(f) with a frequency range of 0.1125Hz≤f≤0.2631 Hz, i.e., the SWIM L2 level one-dimensional wave spectrum, to calculate the spectral energy level E of the box on one side of the nadir orbit according to the following formula. Take the average of the values ​​on both sides of each box orbit to obtain the spectral energy level corresponding to the box: : ; , ; in: It represents the average value of the negative fourth power of the 15th to 32nd spectral density detection frequency of the spectrometer's one-dimensional wave spectrum. represents the average value of the 15th to 32nd spectral density of the spectrometer's one-dimensional wave spectrum, and the subscript j represents the jth spectral density value of the spectrometer's one-dimensional wave spectrum.

[0027] Step 6: Normalize the backscatter coefficient value of the nadir point and the average mean square slope of the sea surface corresponding to the box where it is located , estimated wave spectrum energy level Normalize according to the following formula: ; in: represents the initial nadir normalized backscatter coefficient value, represents the normalized backscatter coefficient of the nadir point after normalization, represents the normalized mean sea surface mean square slope, Represents the normalized wave spectrum energy level.

[0028] Step 7: Calculate the sea surface wind speed at the nadir point according to the following formula: ; ; ; ; ; ; ; ;

[0029] in: represents the input matrix, and represents the intermediate variable matrix, represents the output matrix, represents the input weight matrix, represents the output weight matrix, represents the input bias matrix, represents the output bias matrix, Represents the output nadir sea surface wind speed.

[0030] This embodiment takes the wind speed in January 2024 as an example. Figure 2 This is a scatter plot comparison of the wind speed calculated by the algorithm of the present invention and the time and longitude and latitude of the ERA5 reanalysis data set. Figure 3This is a scatter plot comparison of the wind speed given by the latest version of SWIM (v7.0) and the fifth generation of the European Centre for Medium-Range Weather Forecasts reanalysis product ERA5 data set after matching the time and longitude. The upper left corner is marked with the average deviation, standard deviation, root mean square error, correlation coefficient and number of samples of the two. By comparison, it can be concluded that the average deviation, standard deviation and root mean square error of the wind speed calculated by the algorithm of the present invention and the ERA5 wind speed are better than the wind speed given by the latest version of SWIM (v7.0), indicating that the nadir sea surface wind speed calculated by the algorithm of the present invention is more accurate.

[0031] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

Claims

1. A method for inverting the wind speed at the nadir point of an ocean wave spectrometer, characterized in that: The specific steps include: Step 1: Read the parameters used for nadir sea surface wind speed inversion in SWIM L2 product data; including: surface type identifier, nadir rainfall identifier, nadir box index, nadir normalized backscatter coefficient, nadir normalized backscatter coefficient identifier, linear fitting coefficient of NRCS profile, wave spectrum number, one-dimensional wave spectrum, land surface proportion within the observation range, and sea ice surface proportion within the observation range; Step 2: Filter the read data and retain valid data; Step 3: Based on the nadir box index, the nadir point normalized backscatter coefficient corresponds to the box where it is located; Step 4: Calculate the mean sea surface mean square slope based on the linear fitting coefficient of the NRCS profile ; Step 5: Calculate the spectrum energy level based on the one-dimensional wave spectrum ; Step 6: Normalize the nadir point standardized backscatter coefficient, the average sea surface mean square slope of the box where the nadir point standardized backscatter coefficient is located, and the spectral energy level of the box where the nadir point standardized backscatter coefficient is located; calculate the nadir point sea surface wind speed based on the normalized data.

2. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 1, characterized in that: The step 2 specifically includes: performing data quality control according to the surface type identifier, the nadir rainfall identifier, the nadir normalized backscatter coefficient and the wave spectrum product quality identifier, removing non-ocean surface data and data with quality lower than a preset value, and the wave spectrum product quality identifier includes the proportion of land surface within the observation range and the proportion of sea ice surface within the observation range.

3. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 1, characterized in that: The step 4 is calculated by the following method : Step 4.1: Calculate the mean square slope of the sea surface along the radar observation direction : ； Where B represents the slope of the linear fitting coefficient of the NRCS profile, and e represents the exponent; Step 4.2: Calculate the average sea surface mean square slope of all radar observation directions on one side of the nadir track for each box, and use the average value as the average sea surface mean square slope on one side of the box. Then calculate the average of the average sea surface mean square slopes of the box on both sides of the track to obtain the average sea surface mean square slope corresponding to the box. .

4. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 1, characterized in that: The step 5 adopts the following method: first convert the slope spectrum in the one-dimensional wave spectrum into a height spectrum, then convert the height spectrum into a frequency spectrum F(f) according to the deep water wave dispersion relationship, and calculate the spectral energy level E of the box on the nadir orbit side based on the frequency spectrum according to the following formula: ； Among them, j represents the jth spectral density value of the one-dimensional wave spectrum, f represents the frequency, It represents the average value of the negative fourth power of the 15th to 32nd spectral density detection frequency of the spectrometer's one-dimensional wave spectrum. It represents the average value of the 15th to 32nd spectral density of the one-dimensional wave spectrum of the spectrometer. and The expression is as follows: ， ； Calculate the average of the spectral energy levels of the box on both sides of the nadir orbit .

5. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 4, characterized in that: The slope spectrum S(k) is converted to the height spectrum F(k) according to the following formula: ； Among them, k represents the wave number corresponding to the frequency; The wave number k is converted into frequency f according to the following formula: ； The height spectrum is converted to the frequency spectrum according to the following formula: ； in, Represents the acceleration due to gravity.

6. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 1, characterized in that: Step 6: Perform normalization according to the following formula: ； in, represents the normalized backscatter coefficient of the nadir point after normalization. represents the initial nadir normalized backscatter coefficient value, represents the normalized mean square slope of the sea surface, represents the normalized spectral energy level.

7. The method for inverting the nadir wind speed of an ocean wave spectrometer according to claim 1, characterized in that: Step 6: Calculate the sea surface wind speed at the nadir point according to the following formula: : ； in, is the intermediate variable matrix, and the expression is as follows: ； in, represents the weight matrix, represents the intermediate variable matrix, is the bias matrix, ; and The expression is as follows: ； ； in, represents the weight matrix, represents the bias matrix, represents the input matrix, , and The expression is as follows: ； ； ； in, represents the normalized backscatter coefficient of the nadir point after normalization. represents the normalized mean square slope of the sea surface, represents the normalized spectral energy level.

Images