Sea surface wind vector retrieval off the Yangtze Estuary based on Sentinel-1A
-
摘要: 选取长江口外近岸海域，以欧空局发射的Sentinel-1A为SAR图像数据源，利用多级小波变换和傅里叶变换获得高精度风向信息，再利用基于GMF的3种CMOD系列模型反演风速，最后将反演结果与同时相的ECMWF模式风场数据以及五组实测数据进行对比.结果表明，除个别包含复杂纹理特征的SAR子图像以外，SAR风场反演精度整体优于ECMWF模式风场.经过多级小波变换后的风向反演结果更优，均方根误差分别为31.6°，29.7°，23.5°.经过二级小波变换之后的整景SAR图像结合风向信息代入到CMOD系列模型中反演得到的风速精度最优，均方根误差控制在0.8 m/s.比较适合长江口外近海海域的是MOD-IFR2和CMOD4，均方根误差分别为1.08 m/s和1.05 m/s.Abstract: This paper selected SAR images in the Yangtze Estuary from Sentinel-1A which was launched by ESA. Firstly, it extracts high-precise wind direction using multilevel wavelet transform and fourier transform. Secondly, wind speed information are obtained through three kinds of CMOD models of GMF, and compare with five sets of in-situ data and as well as wind vector from ECMWF. The results indicats that except for some sub-images containing complex texture, the precision of SAR-derived wind field is better than that of ECMWF. The multiscale wavelet transform is conducive to wind field retrieval (RMSE of wind direction are 31.6°, 29.7°, 23.5° respectively), especially the wind speed retrieval after the second level wavelet (RMSE of wind speed is 0.8 m/s). Among CMOD models, CMOD-IFR2 and CMOD4 are suitable for coastal area of the Yangtze Estuary (RMSE of wind speedis 1.08 m/s and 1.05 m/s respectively).
-
Key words:
- synthetic aperture radar /
- wind vector retrieval /
- wavelet transform /
- fourier transform /
- CMOD models
-
图 2 Sentinel-1A图像反演海表风矢量场流程图
Fig. 2 Flowchart of sea surface wind vector field inversion based on Sentinel-1A
图 3 预处理后的2016年6月7日长江口近岸Sentinel-1A图像
Fig. 3 Preprocessed Sentinel-1A image located in Yangtze Estuary offshore on June 7, 2016
图 5 3级小波变化分解后的低频分量和高频分量
Fig. 5 Multiscale components of wavelet transform decomposition
图 6 基于CMOD-ifr2/4/5模型反演结果与验证数据风矢量对比图
Fig. 6 The inversion results of wind vector based on CMOD-ifr2/4/5 models compared with the validation data
图 7 长江口近岸风场反演结果图
注: 分辨率为0.25$^{\circ}\times$0.25$^{\circ}$
Fig. 7 Retrieved sea surface wind field off the Yangtze Estuary
表 1 实测数据
Tab. 1 In-situ data
实测点 北纬/($^{\circ}$) 东经/($^{\circ}$) 风向/($^{\circ}$) 风速/(m$\cdot $s$^{-1}$) S1 31.23 122.22 92.50 7.00 S2 31.00 122.17 145.07 6.90 S3 31.33 122.53 100.50 6.10 S4 31.00 122.53 180.50 8.25 S5 31.42 122.24 125.50 5.75 
下载: 导出CSV
表 2 试验区域SAR提取风向信息与验证数据结果
Tab. 2 S1-S5Wind direction data from SAR and validation data
风向/(°) S1 S2 S3 S4 S5 AAE RMSE In-Situ 92.5 145.1 100.5 180.5 125.5 ECMWF 123.8 119.3 124.1 117.6 126.5 28.896 1 35.080 2 FFT 99.6 231.6 108.0 201.6 149.3 29.207 7 41.487 6 A1+FFT 99.1 207.1 90.6 206.1 144.2 24.565 7 31.604 3 A2+FFT 87.1 200.4 81.2 197.6 151.3 24.585 7 29.735 3 A3+FFT 89.6 191.0 103.1 178.3 100.2 15.785 7 23.535 2
下载: 导出CSV
表 3 实验区域SAR反演风向信息与验证数据结果
Tab. 3 S1-S5 Wind Speed data from SAR and validation data
S1 S2 S3 S4 S5 AAE RMSE In-Situ 7.0 6.9 6.1 8.3 5.8 ECMWF 5.1 5.2 4.8 5.1 4.9 1.788 5 1.788 5 1.953 0 1.953 0 FFT+CMOD4 8.6 7.5 7.3 5.7 5.9 1.220 0 1.475 5 FFT+CMOD-IFR2 9.7 7.6 7.0 6.1 6.0 1.340 0 1.520 0 1.629 4 1.776 5 FFT+CMOD5 5.0 5.2 5.0 4.4 4.4 2.000 0 2.224 6 A1+FFT+CMOD4 8.5 6.4 5.5 6.2 5.4 1.000 0 1.198 7 A1+FFT+CMOD-IFR2 9.5 6.5 6.4 6.5 5.5 1.040 0 1.220 0 1.387 4 1.491 3 A1+FFT+CMOD5 6.2 4.7 5.0 5.0 5.0 1.620 0 1.887 6 A2+FFT+CMOD4 6.6 6.0 5.7 9.4 5.4 0.640 0 0.717 6 A2+FFT+CMOD-IFR2 7.5 6.3 6.0 7.5 6.0 0.440 0 0.720 0 0.499 0 0.798 7 A2+FFT+CMOD5 7.5 5.0 5.0 9.4 5.0 1.080 0 1.179 4 A3+FFT+CMOD4 7.8 6.4 7.3 9.6 5.6 0.800 0 0.913 8 A3+FFT+CMOD-IFR2 6.2 6.1 7.0 8.8 6.0 0.660 0 0.873 3 0.700 7 0.965 4 A3+FFT+CMOD5 7.5 5.0 5.0 9.9 5.1 1.160 0 1.281 8 CMOD4 0.92 1.08 CMOD-IFR2 0.87 1.05 CMOD5 1.47 1.64
下载: 导出CSV
-
[1] JOHANNESSEN J A. Coastal observing systems:The role of synthetic aperture radar[J]. Johns Hopkins Apl Technical Digest, 2000, 21(1):41-48. http://d.wanfangdata.com.cn/OAPaper/oai_doaj-articles_dee14bdcbc2020959c9107b3efe2e774 [2] GERLING T W. Structure of the surface wind field from the Seasat SAR[J]. Journal of Geophysical Research, 1986:2308-2320. doi: 10.1029/JC091iC02p02308/full [3] FETTERER F M, GINERIS D, WACKERMAN C C, et al. Validating a scatterometer wind algorithm for ERS-1 SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 1998, 36(2):479-492. doi: 10.1109/36.662731 [4] MULLER G, BRUMMER B, ALPERS W, et al. Roll convection within an arctic cold-air outbreak:Interpretation of in situ aircraft measurements and spaceborne SAR imagery by a three-dimensional atmospheric model[J]. Monthly Weather Review, 1999, 127(3):363-380. doi: 10.1175/1520-0493(1999)127<0363:RCWAAC>2.0.CO;2 [5] WACKERMAN C C. Estimating wind vectors from RADARSAT synthetic aperture radar imagery[R]. Verdian ERIM International Report 10032100-1T, prepared for Office of Naval Research for Contract, 2000(00014). [6] HORSTMANN J, KOCH W, LEHNER S, et al. Ocean winds from RADARSAT-1 scan SAR[J]. Canadian Journal of Remote Sensing, 2002, 28(3):524-533. doi: 10.5589/m02-043 [7] HORSTMANN J, KOCH W, LEHNER S, et al. Ocean wind fields retrieved from the advanced synthetic aperture radar aboard ENVISAT[J]. Ocean Dynamics, 2004, 54(6):570-576. doi: 10.1007/s10236-004-0098-3 [8] FICHAUX N, RANCHIN T. Combined extraction of high spatial resolution wind speed and wind direction from SAR images:A new approach using wavelet transform[J]. Canadian Journal of Remote Sensing, 2011, 28(3):510-516. doi: 10.5589/m02-038 [9] DU Y, VACHON P W, WOLFE J, et al. Wind direction estimation from SAR images of the ocean using wavelet analysis[J]. Canadian Journal of Remote Sensing, 2002, 28(3):498-509. doi: 10.5589/m02-029 [10] LEITE G C, USHIZIMA D, MEDEIROS F N, et al. Wavelet analysis for wind fields estimation[J]. Sensors, 2010, 10(6):5994-6016. doi: 10.3390/s100605994 [11] 张毅, 蒋兴伟, 宋清涛, 等. C波段地球物理模式函数对比分析研究[J].遥感信息, 2012(1):31-36. http://d.wanfangdata.com.cn/Periodical/ygxx201201007 [12] LA T V, KHENCHAF A, COMBLET F, et al. Study of wind speed retrievals from Sentinel-1 images using physical models[C]//Antennas and Propagation (EuCAP), 201610th European Conference on. IEEE, 2016:1-4. http://ieeexplore.ieee.org/document/7481445/ [13] 孔毅, 赵现斌, 艾未华, 等.基于墨西哥帽小波变换的机载SAR海面风场反演[J].解放军理工大学学报(自然科学版), 2011(3):301-306. http://d.wanfangdata.com.cn/Periodical/jfjlgdxxb201103018 [14] HE Y, ZOU Q. Validation of wind vector retrieval from ENVISAT ASAR images[C]//Geoscience and Remote Sensing Symposium, 2004, IGARSS'04. IEEE, 2004:3184-3187 vol.5. http://www.researchgate.net/publication/224754558_Validation_of_wind_vector_retrieval_from_ENVISAT_ASAR_images [15] 宋贵霆, 侯一筠, 何宜军. Comparison of two wind algorithms of ENVISAT ASAR at high wind[J].中国海洋湖沼学报(英文版), 2006, 24(1):92-96. http://kns.cnki.net/KCMS/detail/detail.aspx?filename=hyfw200601015&dbname=CJFD&dbcode=CJFQ [16] MONALDO F, JACKSON C, PICHEL W, et al. Early validation of operational SAR wind retrievals from Sentinel-1A[C]//Geoscience and Remote Sensing Symposium. IEEE, 2015:1223-1226. http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7325993 [17] PICHEL W G, MONALDO F M, JACKSON C, et al. NOAA operational SAR winds:Current status and plans for Sentinel-1A[C]//Geoscience and Remote Sensing Symposium. IEEE, 2015:4916-4919. http://ieeexplore.ieee.org/document/7326934/ [18] 范开国, 黄韦艮, 常俊芳, 等.NCEP/QSCAT混合风向用于SAR图像反演高分辨率海面风速[J].遥感技术与应用, 2010, 25(6):873-876. doi: 10.11873/j.issn.1004-0323.2010.6.873 [19] ALPERS W, BRÜMMER B. Atmospheric boundary layer rolls observed by the synthetic aperture radar aboard the ERS-1 satellite[J]. Journal of Geophysical Research Oceans, 1994, 99(C6):12613-12621. doi: 10.1029/94JC00421 [20] 张毅, 蒋兴伟, 林明森, 等.基于小波分析的近岸海面风场反演研究[J].高技术通讯, 2011, 21(10):1056-1061. doi: 10.3772/j.issn.1002-0470.2011.10.010 [21] 张雷, 石汉青, 杜华栋, 等.用小波分析反演星载SAR图像海面风向[J].遥感学报, 2013, 18(1):215-230. http://d.wanfangdata.com.cn/Periodical/ygxb201401015 [22] 杨劲松, 黄韦艮, 周长宝, 等.合成孔径雷达图像的近岸海面风场反演[J].遥感学报, 2001, 5(1):13-16. doi: 10.11834/jrs.20010103 [23] 程玉鑫, 艾未华, 孔毅, 等.基于影像纹理特征和外部风向的星载SAR海面风场反演研究[J].海洋科学, 2015, 39(12):157-164. doi: 10.11759/hykx20141027002 [24] LIN H, XU Q, ZHENG Q. An overview on SAR measurements of sea surface wind[J]. Progress in Natural Science, 2008, 18(8):913-919. doi: 10.1016/j.pnsc.2008.03.008 [25] WACKERMAN C C, RUFENACH C L, SHUCHMAN R A, et al. Wind vector retrieval using ERS-1 synthetic aperture radar imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 1996, 34(6):1343-1352. doi: 10.1109/36.544558 [26] PORTABELLA M, STOFFELEN A, JOHANNESSEN J A. Toward an optimal inversion method for synthetic aperture radar wind retrieval[J]. Journal of Geophysical Research:Oceans, 2002, 107(C8):1-13. doi: 10.1029/2001JC000925/full [27] 王珂, 洪峻, 张问一, 等. SAR反演邻近岸海面风场方法[J].电子与信息学报, 2013, 35(8):1800-1805. http://d.wanfangdata.com.cn/Periodical/dzkxxk201308004 -
点击查看大图
图(7) / 表(3)
计量
- 文章访问数: 221
- HTML全文浏览量: 81
- PDF下载量: 360
- 被引次数: 0