On the wind inversion characteristics of China-France Oceanography Satellite microwave scatterometer
-
摘要: 2018年10月发射的中法海洋卫星散射计(CSCAT)是国际上首个扇形波束旋转扫描微波散射计。本文以最大似然估计风场反演算法为基线，详细分析了中法海洋卫星微波散射计海面风场反演代价函数的残差特性，重点研究了新的观测几何对风场反演残差以及风场质量的影响，并建立了风场模糊解的似然概率模型函数。结果表明，CSCAT风场反演的残差特性随风矢量单元在刈幅交轨方向位置的变化而变化，模糊解似然概率模型函数的指数分布在−0.4～−1.8之间。分析结果为CSCAT风场质量控制和二维变分分析去模糊算法的精细化调整提供了重要的参考。Abstract: China-France Oceanography Satellite scatterometer (CSCAT) is the first rotating fan beam scatterometer internationally, which was flown onboard China-France Oceanography Satellite (CFOSAT) on October 2018. Based on the maximum likelihood estimation wind inversion algorithm, the residual characteristics of the CSCAT sea surface wind inversion cost function in detail, focuses on the influence of the new observation geometry on the wind inversion residual and wind quality is analyzed in this article, we establish the likelihood probability model function of the ambiguous solutions. The results show that the residual characteristic of the CSCAT wind inversion varies with the position of the wind vector cell (WVC) across the swath. The exponential distribution of the ambiguous solution likelihood probability model function is between −0.4 and −1.8. The results provide an important reference for the quality control of CSCAT and the refinement adjustment of the two dimensional variational ambiguity removal algorithm.
-
图 5 平均MLE随风速和WVC列数的变化（a）及MLE标准差随风速和WVC列数的变化（b）
Fig. 5 Mean MLE versus wind speed and WVC number (a) and MLE standard deriation versus wind speed and WVC number (b)
图 6 CSCAT风速、风向相对于ECMWF风速、风向的偏差随平均MLE变化的曲线
a. 风速偏差；b. 风向偏差；不同线型的曲线表示不同的WVC列数
Fig. 6 The variation curve of CSCAT wind speed and direction bias relative to ECMWF wind speed and direction with mean MLE
a. Biases of wind speed; b. biases of wind direction; lines in different styles are for different WVC number
图 7 CSCAT风速、风向相对于ECMWF风速、风向的标准差随平均MLE变化的曲线
a. 风速标准差；b. 风向标准差；不同线型的曲线表示不同的WVC列数
Fig. 7 The variation curve of CSCAT wind speed and direction standard deviation relative to ECMWF wind speed and direction with mean MLE
a. Standard deviation of wind speed differences; b. standard deviation of wind direction differences; lines in different styles are for different WVC number
图 8 不同节点下（
$ {Rn}_{2}-{Rn}_{1} $ ）值与$ p\left({Rn}_{2}\right)/p\left({Rn}_{1}\right) $ 函数关系的指数拟合（$ {Rn}_{1} $ =0.1）Fig. 8 The exponential fit to ratio
$ {Rn}_{2} $ and$ {Rn}_{1} $ as a founction of ($ {Rn}_{2}-{Rn}_{1} $ ) value for different WVC number（$ {Rn}_{1} $ =0.1）
图 9 不同
$Rn_1 $ 取值下$ p\left({Rn}_{2}\right)/p\left({Rn}_{1}\right) $ 与（$ {Rn}_{2}-{Rn}_{1} $ ）值函数关系的指数拟合（第8列WVC）Fig. 9 The exponential fit to ratio of
$ {Rn}_{2} $ and$ {Rn}_{1} $ as a founction of ($ {Rn}_{2}-{Rn}_{1} $ ) value for different$ {Rn}_{1} $ (WVC number 8)
图 10
$ {Rn}_{1} $ 分布直方图（节点数为8）Fig. 10 The distribution histogram of
$ {Rn}_{1} $ (the number of node is 8)
图 11 概率模型函数的指数及系数随节点的变化
Fig. 11 The exponents and coefficients of the probabilistic model versus node number
图 12 改进前概率模型函数反演的CSCAT风场与浮标测量风场
Fig. 12 CSCAT wind filed and buoy wind filed retrieved by the likelihood probability model before improvement
图 13 改进的概率模型函数反演的CSCAT风场与浮标测量风场
Fig. 13 CSCAT wind filed and buoy wind filed retrieved by the improved likelihood probability model
表 1 预测概率/观测概率的分布对比(刈幅远端)
Tab. 1 Distribution comparision of predicted probability/observed probability (far swath)
2个模糊解/% 3个模糊解/% 4个模糊解/% 所有模糊解/% 风单元个数 267 607 117 849 458 349 472 138 第1模糊解 84/81 82/79 83/80 83/80 第2模糊解 16/19 13/16 10/13 14/17 第3模糊解 − 5/5 4/4 3/3 第4模糊解 − − 3/3 0/0 注：−代表未获得概率。 
下载: 导出CSV
表 2 预测概率/观测概率的分布对比(刈幅中间)
Tab. 2 Distribution comparision of predicted probability/observed probability (sweet swath)
2个模糊解/% 3个模糊解/% 4个模糊解/% 所有模糊解/% 风单元个数 514 639 345 693 273 984 1 134 316 第1模糊解 90/88 78/82 80/83 84/85 第2模糊解 10/12 16/14 14/13 13/13 第3模糊解 − 6/4 4/3 3/2 第4模糊解 − − 2/1 0/0 注：−代表未获得概率。
下载: 导出CSV
表 3 预测概率/观测概率的分布对比 (星下点区域)
Tab. 3 Distribution comparision of predicted probability/observed probability (nadir swath)
2个模糊解/% 3个模糊解/% 4个模糊解/% 所有模糊解/% 风单元个数 162 112 144 373 40 206 346 691 第1模糊解 78/78 56/66 48/56 65/71 第2模糊解 22/22 28/22 23/21 25/22 第3模糊解 − 16/12 17/13 9/6 第4模糊解 − − 12/10 1/1 注：−代表未获得概率。
下载: 导出CSV
-
[1] Liu W T. Progress in scatterometer application[J]. Journal of Oceanography, 2002, 58(1): 121−136. doi: 10.1023/A:1015832919110 [2] 蒋兴伟, 林明森, 张有广, 等. 海洋遥感卫星及应用发展历程与趋势展望[J]. 卫星应用, 2018(5): 10−18. doi: 10.3969/j.issn.1674-9030.2018.05.005Jiang Xingwei, Lin Mingsen, Zhang Youguang, et al. Progress and trend of ocean remote sensing satellites and their applications[J]. Satellite Application, 2018(5): 10−18. doi: 10.3969/j.issn.1674-9030.2018.05.005 [3] Lin Wenming, Portabella M. Toward an improved wind quality control for RapidScat[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(7): 3922−3930. doi: 10.1109/TGRS.2017.2683720 [4] Stoffelen A, Anderson D. Scatterometer data interpretation: measurement space and inversion[J]. Journal of Atmospheric and Oceanic Technology, 1997, 14(6): 1298−1313. doi: 10.1175/1520-0426(1997)014<1298:SDIMSA>2.0.CO;2 [5] Portabella M, Stoffelen A. Characterization of residual information for SeaWinds quality control[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(12): 2747−2759. doi: 10.1109/TGRS.2002.807750 [6] De Vries J C W, Stoffelen A C M. 2D variational ambiguity removal[R]. Netherlands Remote Sensing Board (BCRS), Programme Bureau, Rijkswaterstaat Survey Department, 2000. [7] Liu Jianqiang, Lin Wenming, Dong Xiaolong, et al. First results from the rotating fan beam scatterometer onboard CFOSAT[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(12): 8793−8806. doi: 10.1109/TGRS.2020.2990708 [8] Stiles B W, Dunbar R S. A neural network technique for improving the accuracy of scatterometer winds in rainy conditions[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(8): 3114−3122. doi: 10.1109/TGRS.2010.2049362 [9] Portabella M, Stoffelen A. A comparison of KNMI quality control and JPL rain flag for SeaWinds[J]. Canadian Journal of Remote Sensing, 2002, 28(3): 424−430. doi: 10.5589/m02-040 [10] Lin Wenming, Dong Xiaolong, Portabella M, et al. A perspective on the performance of the CFOSAT rotating fan-beam scatterometer[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(2): 627−639. doi: 10.1109/TGRS.2018.2858852 [11] 解学通, 方裕, 陈晓翔, 等. 基于最大似然估计的海面风场反演算法研究[J]. 地理与地理信息科学, 2005, 21(1): 30−33. doi: 10.3969/j.issn.1672-0504.2005.01.009Xie Xuetong, Fang Yu, Chen Xiaoxiang, et al. Research on numerical wind vector retrieval algorithm based on maximum likelihood estimation[J]. Geography and Geo-Information Science, 2005, 21(1): 30−33. doi: 10.3969/j.issn.1672-0504.2005.01.009 [12] Stoffelen A, Portabella M. On Bayesian scatterometer wind inversion[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(6): 1523−1533. doi: 10.1109/TGRS.2005.862502 [13] Anderson C, Bonekamp H, Duff C. Analysis of ASCAT ocean backscatter measurement noise[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(7): 2449−2457. doi: 10.1109/TGRS.2012.2190739 [14] Portabella M, Stoffelen A. A probabilistic approach for SeaWinds data assimilation[J]. Quarterly Journal of the Royal Meteorological Society, 2010, 130(596): 127−152. [15] Gohil B S, Sarkar A, Agarwal V K. A new algorithm for wind-vector retrieval from scatterometers[J]. IEEE Geoscience and Remote Sensing Letters, 2008, 5(3): 387−391. doi: 10.1109/LGRS.2008.917129 [16] Portabella M, Stoffelen A. Rain detection and quality control of SeaWinds[J]. Journal of Atmospheric and Oceanic Technology, 2001, 18(7): 1171−1183. doi: 10.1175/1520-0426(2001)018<1171:RDAQCO>2.0.CO;2 -
点击查看大图
计量
- 文章访问数: 339
- HTML全文浏览量: 106
- PDF下载量: 28
- 被引次数: 0
