A typhoon wind field retrieval method based on synthetic aperture radar remote sensing image
By constructing a typhoon wind field inversion model based on co-polarization and cross-polarization, the problems of scattering cross section saturation and wind direction discontinuity in wind field reconstruction of SAR remote sensing images were solved, achieving high-precision sea surface wind field reconstruction, especially improving the accuracy of wind speed and wind direction prediction in the typhoon center region.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI OCEAN UNIV
- Filing Date
- 2023-07-31
- Publication Date
- 2026-06-12
AI Technical Summary
In existing technologies, SAR remote sensing images with the same polarization mode suffer from backscatter cross section saturation in sea surface wind field reconstruction and inversion. Furthermore, wind direction changes have a significant impact on the backscatter cross section value, resulting in low wind direction prediction accuracy and discontinuity at the sub-strip edges.
By acquiring spaceborne SAR remote sensing images and actual observation data, and utilizing the relationship between the maximum wind speed and the maximum wind speed radius under co-polarization and cross-polarization, a typhoon wind field inversion model is constructed. Combined with the C-SARMOD algorithm and geophysical model functions, the sea surface wind field is reconstructed and inverted.
High-precision sea surface wind field reconstruction was achieved, especially in the typhoon center region, where the accuracy of wind speed and direction prediction was improved, sub-band edge distortion was reduced, and the integrity of the wind field structure was enhanced.
Smart Images
Figure CN116953708B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of synthetic aperture radar technology, specifically to an algorithm for inverting typhoon wind fields from synthetic aperture radar remote sensing images. Background Technology
[0002] Sea surface wind field is one of the most important parameters in ocean-atmosphere interaction research. Wind has a direct or indirect impact on various movements in the ocean and plays a crucial role in regulating the exchange of heat, matter, and water vapor between the ocean and the atmosphere. Various studies in oceanography and meteorology require sea surface wind field data. Traditional observation methods include buoys, ocean stations, and irregular ship reports. However, due to the extreme weather conditions of typhoons, real-time on-site observation data of the ocean areas they pass through are often lacking. The Northwest Pacific, as one of the regions most severely affected by typhoons, frequently requires satellite remote sensing data for real-time monitoring and forecasting of typhoon paths, intensity, and structure. With the development of remote sensing technology, various passive and active microwave sensors have become important means of acquiring sea surface wind fields, such as radiometers, altimeters, scatterometers, and synthetic aperture radar. Among them, synthetic aperture radar (SAR) can acquire sea surface wind field information with relatively higher spatial resolution. It is more suitable for measuring sea surface wind fields and local wind fields in nearshore areas, islands, and sea areas near ice edges. To a certain extent, it makes up for the shortcomings of scatterometer wind measurement and plays an increasingly important role in typhoon monitoring. It has received much attention in recent years.
[0003] SAR (Surveyorical Reflection) is an active microwave imaging radar. SAR measures the backscattered signal from the sea surface, and after processing, produces a backscattered intensity map, which is an intensity image characterizing sea surface roughness. Sea surface winds, through their interaction with the sea surface, generate surface waves, altering sea surface roughness and causing changes in the radar backscattering coefficient signal. Therefore, by utilizing the correlation between sea surface roughness information and sea surface wind field in SAR images, sea surface wind field information can be retrieved. Geophysical Model Functions (GMFs) are a commonly used method for retrieving sea surface wind fields.
[0004] The 21st century is the century of the ocean. With the continuous development of ocean exploration technologies, traditional wind measurement methods such as shore-based observation stations, ships, buoys, and spaceborne microwave scatterometers are no longer sufficient to meet the data requirements for certain high-resolution sea surface wind fields. In marine remote sensing technology, Synthetic Aperture Radar (SAR) has certain advantages. SAR is an active microwave imaging radar that transmits microwaves of a specific frequency and measures the amplitude and phase information of their backscattered signals to obtain images of the sea surface backscattering intensity. These images have very high resolution, reaching the level of several meters. Furthermore, SAR is highly sensitive to changes in sea surface roughness, providing a wealth of ocean dynamic information on subtle spatial variations in the sea surface, such as sea surface wind fields, sea surface waves, internal waves, ocean currents, sea ice, wakes of ships on the sea surface, and oil slicks. Simultaneously, as a microwave imaging radar, SAR can observe the sea surface at any time and in any weather condition. Therefore, SAR is an all-weather, all-time, high-resolution imaging radar for ocean observation.
[0005] In existing technologies, SAR can continuously acquire high-resolution images of large areas of the ocean surface around the clock and in all weather conditions, which is highly advantageous for monitoring sea surface wind fields. For example, the strip-type remote sensing images from the next-generation C-band Sentinel-1 SAR satellite launched by the European Space Agency (ESA) in 2014 have a single pixel width of up to 10 meters and a swath width of up to 80 kilometers. Currently, Sentinel-1 SAR can provide four dual polarization modes: vertical-vertical (VV polarization), horizontal-horizontal (HH polarization), horizontal-vertical (HV polarization), and vertical-horizontal (VH polarization). The sea surface wind field inversion technology CMODs (Co-polarization of C-band SAR, i.e., VV polarization and HH polarization) is fully mature, and the geophysical model functions (GMFs) of CMODs have been widely used over the past few decades, such as CMOD4, CMOD-IFR2, CMOD5, the neutral wind inversion algorithm CMOD5N, and the new GMF C-SARMOD, etc. To date, CMODs, which describe the complex relationship between the SAR normalized radar cross section (NRCS) and the wind speed at 10 meters above the sea surface, have been successfully applied to the sea surface wind field inversion of C-band SAR remote sensing images. A recent study used the ASCAT scatterometer on the Metop-A and Metop-B satellites to numerically verify the sea surface wind field inversion results of C-band Sentinel-1SAR remote sensing images. The comparison results show that the CMOD5N and CMOD4+PR inversion algorithms are applicable to the inversion of sea surface wind fields in VV polarization and HH polarization modes, respectively. Summary of the Invention
[0006] Previous research has shown that using cross-polarization SAR remote sensing images for sea surface wind field inversion has the following advantages: There is a strong linear relationship between the backscatter cross section (RSS) of cross-polarization SAR images and wind speed; that is, the RSS value increases with increasing wind speed. In contrast, cross-polarization SAR images are less dependent on wind direction, meaning that changes in wind direction have a relatively smaller impact on the RSS value. In other words, cross-polarization SAR images can more reliably invert wind speed in sea surface wind fields, while offering relatively high accuracy in wind direction prediction. The RSS value of co-polarization SAR images typically increases gradually with increasing wind speed, reaching a saturation point when the wind speed reaches 25 m / s.
[0007] To address the challenges of existing sea surface wind field reconstruction and inversion algorithms for SAR remote sensing images using the same polarization method, which require leveraging the strong linear relationship between the backscatter cross section and wind speed to eliminate the saturation problem of the backscatter cross section at a certain wind speed, and also needing to utilize the absence of sub-band edges in the same polarization method to avoid severe discontinuities in wind inversion, this paper proposes a typhoon wind field inversion algorithm for synthetic aperture radar (SAR) remote sensing images. The technical solution is as follows:
[0008] A method for reconstructing and inverting sea surface wind fields based on synthetic aperture radar remote sensing images, characterized in that the method includes:
[0009] Acquire raw spaceborne SAR remote sensing images, actual observed SFMR, and SMAP wind field data;
[0010] The relationship between wind speed and maximum wind speed radius is explored using the maximum wind speed of the same polarization, the maximum wind speed of the cross polarization, and the maximum wind speed radius.
[0011] Based on the research results, an inversion model of the typhoon wind field was constructed;
[0012] The accuracy of the constructed inversion model was verified, and the inversion model was used to reconstruct and invert the sea surface wind field.
[0013] A further improvement of this invention lies in acquiring raw spaceborne SAR remote sensing images and actual observed SMAP wind field data, specifically including:
[0014] Multiple spaceborne SAR remote sensing images were collected; SFMR wind field data with a spatial resolution of 0.01° were collected; and SMAP (Soil Moisture Active Passive) reanalysis wind field data with an interval of 1 hour and a resolution of 0.1° were collected.
[0015] A further improvement of this invention lies in exploring the relationship between wind speed and maximum wind speed radius using the maximum wind speed of the same polarization, the maximum wind speed of cross-polarization, and the maximum wind speed radius, specifically including:
[0016] After radiometric calibration, the SAR remote sensing image to be inverted is divided into multiple sub-images. The wind speed of the same polarization wind field is obtained using the C-SARMOD algorithm, and the wind speed under cross-polarization of IW and EW is calculated by the geophysical model functions of S1IW.NR and S1EW.NR, respectively. The relationship between the ratio of the same polarization wind speed and the maximum cross-polarization wind speed and the maximum wind speed radius is determined. In the formula, we use the relationship between the ratio of the maximum cross-polarization wind speed and the maximum same polarization wind speed.
[0017] A further improvement of this invention lies in constructing an inversion model of the typhoon wind field based on the research results, specifically including:
[0018] The following steps are used to construct an inversion method for reconstructing sea surface wind fields from SAR remote sensing images:
[0019] (1) The wind speed of the same polarization is calculated using the same polarization physical model function, as shown in the following formula:
[0020]
[0021] Where σ0 is the backscattering cross section in VV polarization. This refers to the wind direction relative to the satellite's flight path. Matrix B represents the wind speed U above 0m from the sea surface. 10 and angle of incidence;
[0022] (2) The cross-polarized wind speed is calculated using geophysical model functions, where the algorithms for IW and EW are shown below;
[0023] S1EW.NR:
[0024]
[0025] Among them U 10 For wind speed, The backscattering coefficient of VH polarization radar.
[0026] Additional correction is required at low incident angles, as shown below.
[0027]
[0028] S1IW.NR:
[0029] 3. At wind speed U 10 When ≤30m / s:
[0030]
[0031] 4. At wind speed U 10 When >30m / s:
[0032]
[0033] (3) Estimate the identification value of the TC (Tropical Cyclone) center based on the SAR inversion wind field from the VH polarization image; calculate the TC parameters based on the SAR wind speed inversion, including: the maximum wind speed of VV polarization, denoted as Maxwind_VV; the maximum wind speed of VH polarization, denoted as Maxwind_VH; and the maximum wind speed radius R. max_ VH; based on VV polarization SAR wind speed U10 The typhoon wind field was reconstructed using VH polarized typhoon characteristic parameters. This wind field has the characteristics of small sub-strip edge distortion and high accuracy in medium and low winds, as shown in the following formula.
[0034]
[0035] In the above formula,
[0036]
[0037] r is the distance correction parameter.
[0038] A further improvement of this invention lies in verifying the accuracy of the algorithm for constructing the wind field, specifically including:
[0039] The entire SAR image was divided into several sub-scenes: IW mode (256×256 pixels, ~3 km) and EW mode (128×128 pixels, ~5 km). Reconstruction methods were applied to all sub-scenes. Compared with VV-polarized and VH-polarized SAR wind fields, the reconstructed wind field showed better structural integrity. In particular, the improvement in the discontinuity of the VH-polarized SAR wind field is significant. The retrieval results of the reconstructed wind field were compared with SFMR observations and SMAP products. The results showed that the reconstructed wind field had higher accuracy. In this sense, we believe that the wind retrieved from the reconstructed SAR is quite reliable for typhoon research.
[0040] The beneficial effects of the technical solution provided by this invention are:
[0041] This invention discloses a method for reconstructing and inverting sea surface wind fields based on synthetic aperture radar (SAR) remote sensing images. It utilizes spaceborne SAR to acquire remote sensing images, performs radiometric calibration on these images, and obtains their backscattering cross section and incident angle. Combining the advantages of co-polarized and cross-polarized SAR wind speed algorithms, this invention proposes a novel typhoon wind field inversion algorithm. Attached Figure Description
[0042] To more clearly illustrate the technical solution of this invention, the accompanying drawings required in the algorithm development description will be briefly introduced below.
[0043] Figure 1 This is a flowchart of a typhoon wind field inversion method based on synthetic aperture radar remote sensing images according to the present invention.
[0044] Figure 2 a and 2b are quick views of the same polarization and cross-polarization normalized radar cross section (NRCS) over the typhoon, respectively.
[0045] Figure 3 (a) is the result of the inversion of the VV polarization image. Figure 3(b) is the inversion result of the VH polarization image;
[0046] Figure 4 This is a wind field map reconstructed from a SAR wind field;
[0047] Figure 5 This is a comparison chart of the reconstructed wind field and SMAP. Detailed Implementation
[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.
[0049] This invention provides a method for reconstructing and inverting sea surface wind fields based on synthetic aperture radar (SAR) remote sensing images. In this invention, the wind speed inversion model (S1IW.NR) after denoising the S-1IW image mode uses VH-polarized GMF to invert typhoon wind speeds. The impact of the differences between VV-polarized NRCS and GMF CMOD5.N simulations under VH-polarized wind speed conditions on rainfall rate was investigated. Based on this, a practical SAR rainfall rate inversion algorithm is proposed. This algorithm was applied to all images and validated with products from the Global Precipitation Measurement (GPM) mission. Specifically:
[0050] 101: Acquire raw spaceborne SAR images, as well as actual observed SFMR and SMAP (Soil Moisture Active Passive) wind field data.
[0051] Specifically, 30 S-1 images acquired between 2016 and 2021 in interferometric width (IW) and ultrawide (EW) modes with a pixel count of 40 μm were available for this work (as shown in Table 1 below). Of these images, spatial coverage was observed in 13 images using NOAA's SFMR (full-frequency microwave radiometer) data. Figure 2 a and 2b show quick views of the same-polarization and cross-polarization normalized radar cross section (NRCS) over the typhoon, respectively, where the red rectangles represent the aircraft's trajectory.
[0052] Table 1: Information on Synthetic Aperture Radar (SAR) Images and Corresponding Tropical Cyclones
[0053]
[0054] 102: Exploring the relationship between the maximum wind speed and the radius of maximum wind speed using co-polarized and cross-polarized maximum wind speeds.
[0055] Specifically, the inversion results of the VV polarization image are as follows: Figure 3 As shown in (a), it can be seen from the figure that the wind speed of VV did not reach the actual typhoon wind speed; the inversion result of the VH polarization image is as follows. Figure 3As shown in (b), it can be seen from the figure that the wind speed of VH has obvious sub-band edge problems.
[0056] 103: Based on the research results, an inversion algorithm for reconstructing the sea surface wind field was constructed.
[0057] Specifically, the location of the typhoon's eye center was estimated based on the SAR inversion wind field from the cross-polarization image. Furthermore, two typhoon parameters were calculated from the SAR: the cross-polarization maximum wind speed (Maxwind_VV represents the maximum wind speed with the same polarization, Maxwind_VH represents the cross-polarization maximum wind speed, and Rmax_VH represents the maximum wind radius). Figure 4 The image shows the wind field after SAR wind field reconstruction in this embodiment.
[0058] 104: Verify the accuracy of the inversion algorithm.
[0059] from Figure 5 As can be seen, the RMSE of the reconstructed wind field compared with the SMAP is 4.65, and the correlation coefficient is 0.94, proving the reliability of the reconstructed wind field. The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make many modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.
Claims
1. A method for reconstructing and inverting sea surface wind fields based on synthetic aperture radar remote sensing images, characterized in that, The method includes: Acquire raw spaceborne SAR remote sensing images, actual observed SFMR, and SMAP wind field data; The relationship between wind speed and maximum wind speed radius is explored using the maximum wind speed of the same polarization, the maximum wind speed of the cross polarization, and the maximum wind speed radius. Based on the research results, an inversion model of the typhoon wind field was constructed; Verify the accuracy of the constructed inversion model and use the inversion model to invert the typhoon wind field; The relationship between wind speed and maximum wind speed radius is explored using the maximum wind speed with the same polarization, the maximum wind speed with cross polarization, and the maximum wind speed radius. Specifically, this includes: After radiometric calibration, the SAR remote sensing image to be inverted is divided into multiple sub-images. The wind speed of the same polarization wind field is obtained using the C-SARMOD algorithm, and the wind speed under cross-polarization of IW and EW is calculated by the geophysical model functions of S1IW.NR and S1EW.NR, respectively. The relationship between the ratio of the same polarization wind speed and the maximum cross-polarization wind speed and the maximum wind speed radius is determined.
2. The method according to claim 1, characterized in that, Acquire raw spaceborne SAR remote sensing images and actual observed SMAP wind field data, specifically including: Multiple spaceborne SAR remote sensing images were collected; SFMR wind field data with a spatial resolution of 0.01° were measured by SFMR; and SMAP reanalysis wind field data with a resolution of 0.1° were collected.
3. The method according to claim 1, characterized in that, Based on the research results, an inversion model for typhoon wind fields was constructed, specifically including: constructing an inversion model for reconstructing sea surface wind fields from SAR remote sensing images according to the following steps: (1) The wind speed of the same polarization is calculated using the same polarization physical model function, as shown in the following formula: ； in, It is the backscattering cross section in VV polarization. Matrix B represents the wind direction relative to the satellite's flight path, while matrix B represents the wind speed U above 10 meters above the sea surface. 10 and angle of incidence The function that determines; (2) The cross-polarized wind speed is calculated using geophysical model functions, where the algorithms for IW and EW are shown below; S1EW.NR: ； in For wind speed, The backscattering coefficient of VH polarization radar. Additional correction is required at low incident angles, as shown below. = +0.5sin(90.24θ+121.01) ; S1IW.NR: At wind speed U 10 When ≤30 m / s: ； At wind speed U 10 When >30 m / s: ； (3) Estimate the identification value of the TC center based on the SAR inversion wind field from the VH polarization image; calculate the TC parameters based on the SAR wind speed inversion, including: the maximum wind speed of VV polarization, denoted as Maxwind_VV; the maximum wind speed of VH polarization, denoted as Maxwind_VH; and the maximum wind speed radius R. max_ VH; Wind speed based on VV polarization SAR The typhoon wind field was reconstructed using VH polarized typhoon characteristic parameters. This wind field has the characteristics of small sub-strip edge distortion and high accuracy in medium and low winds, as shown in the following formula. ； In the above formula, = ； These are the distance correction parameters.