星载GNSS-R融冰期海冰密集度反演研究

王玥; 谢涛; 李建; 张雪红; 白淑英; 王明华

星载GNSS-R融冰期海冰密集度反演研究

王玥^1,,
谢涛^{1, 2, 3, 4, ,},
李建^{1, 3, 4},
张雪红^{1, 3, 4},
白淑英^{1, 3, 4},
王明华^{1, 3, 4}

1.
南京信息工程大学遥感与测绘工程学院，江苏南京 210044
2.
青岛海洋科技中心区域海洋动力学与数值模拟功能实验室，青岛山东 266200
3.
自然资源部遥感导航一体化应用工程技术创新中心，江苏南京 210044
4.
江苏省协同精密导航定位与智能应用工程研究中心，江苏南京 210044

基金项目: 国家自然科学基金项目(42176180)；国家重点研发计划项目(2021YFC2803302)。

详细信息

作者简介:
王玥（1998—），女，江苏省南通市人，主要从事GNSS-R海洋遥感研究。E-mail：wy2805@outlook.com

通讯作者:
谢涛(1973—)，男，湖南省张家界市人，教授，主要从事海洋微波遥感研究。E-mail: xietao@nuist.edu.cn

计量
- 文章访问数: 52
- HTML全文浏览量: 11
- PDF下载量: 10
- 被引次数: 0
出版历程
- 收稿日期: 2023-05-30
- 修回日期: 2023-11-21
- 网络出版日期: 2024-03-14

Sea ice concentration retrieval using spaceborne GNSS-R during the melting period

Wang Yue^1
,,
Xie Tao^{1, 2, 3, 4
, ,},
Li Jian^{1, 3, 4},
Zhang Xuehong^{1, 3, 4},
Bai Shuying^{1, 3, 4},
Wang Minghua^{1, 3, 4}

1.
School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China
2.
Laboratory for Regional Oceanography and Numerical Modeling, Qingdao Marine Science and Technology Center, 266200, Qingdao, Shangdong Province, China
3.
Technology Innovation Center for Integration Applications in Remote Sensing and Navigation, Ministry of Natural Resources, Nanjing 210044, China
4.
Jiangsu Province Engineering Research Center of Collaborative Navigation/Positioning and Smart Application, Nanjing 210044, China

摘要

摘要: 针对北极融冰期的海冰密集度反演，并改善全球导航卫星系统反射测量（GNSS-R）对海水的海冰密集度高估问题，本文提出一种利用机器学习算法生成高时空分辨率的融冰期海冰密集度估算方法，提取GNSS-R时延多普勒图（DDM）的特征参数，并结合海表温度数据建立LightGBM模型，将反演结果与参考海冰密集度值进行相关性分析和评估。本文的模型结果与OSI SAF的海冰密集度产品显示出较好的一致性，相关系数、平均绝对误差和均方根误差分别为0.965、0.061和0.090，该方法能够实现对北极海冰边缘区的海冰密集度高精度估计。
- GNSS-R /
- DDM /
- 融冰期 /
- 海冰密集度 /
- LightGBM /
- 北极
Abstract: In this paper, a high spatial-temporal resolution sea ice concentration estimation method for the Arctic melting season is proposed, aiming to improve the overestimation of sea ice concentration in seawater by the Global Navigation Satellite System-Reflectometry (GNSS-R). The method utilizes machine learning algorithms to extract feature parameters from the Delay Doppler Maps (DDM) obtained through GNSS-R and combines them with sea surface temperature data to establish a LightGBM model. The inversion results are then subjected to correlation analysis and evaluation against reference sea ice concentration values. The model's performance is compared with the sea ice concentration product from OSI SAF, demonstrating good consistency, with correlation coefficient, mean absolute error, and root mean square error being 0.965, 0.061, and 0.090, respectively. This approach enables high-precision estimation of sea ice concentration in the Arctic marginal ice zone.
- GNSS-R /
- DDM /
- melting season /
- sea ice concentration /
- LightGBM /
- Arctic

HTML全文

图 1 0%、50%和90% SIC DDM和对应的归一化积分延迟波形

Fig. 1 DDM for 0%, 50% and 90% SIC and corresponding normalized Integrated Delay Waveforms

下载: 全尺寸图片幻灯片

图 2 LightGBM模型架构

Fig. 2 Framework of the LightGBM model

下载: 全尺寸图片幻灯片

图 3 测试集基于DDM特征参数的反演结果与OSISAF海冰密集度对比散点图（a）和基于DDM特征参数的LightGBM特征重要性排序结果（b）

Fig. 3 Scatter plot of retrieval results based on DDM observables against OSISAF sea ice concentration from the test samples (a) and LightGBM feature importance ranking results based on DDM observables (b)

下载: 全尺寸图片幻灯片

图 4 测试集海冰密集度反演结果与OSISAF海冰密集度对比散点图

a. 基于SST；b. 基于DDM特征参数、SST

Fig. 4 Scatter plot of retrieval results against OSISAF sea ice concentration from the test samples

a. Based on SST; b. based on DDM observables and SST

下载: 全尺寸图片幻灯片

图 5 整个验证集不同海冰密集度区间与反演误差箱线图

a. 基于DDM特征参数的模型；b. 基于DDM特征参数、SST的模型

Fig. 5 Box plots of different sea ice concentration intervals and retrieval error using the entire validation set

a. Based on DDM observables; b based on DDM observables and SST

下载: 全尺寸图片幻灯片

图 6 基于DDM特征参数和SST的反演结果与OSISAF海冰密集度对比散点图

a. 2018年7月20-22日；b. 2018年8月9-10日；c. 2018年8月15-16日

Fig. 6 Scatter plot of retrieval results based on DDM observable against OSISAF sea ice concentration

a. July 20-22, 2018; b. August 9-10, 2018; c. August 15-16, 2018

下载: 全尺寸图片幻灯片

图 7 2018年8月9日模型反演结果与OSISAF海冰密集度地图比较

Fig. 7 Comparison of model retrieval results with OSISAF sea ice concentration map on August 9th, 2018

下载: 全尺寸图片幻灯片

表 1 不同特征参数组合的反演误差统计表

Tab. 1 Error Statistics for Different DDM observables Combinations

序号	特征参数组合	R	MAE	RMSE
1	DDMA + SNR + PP + TES + LES + K + SK (无Γ)	0.867	0.124	0.176
2	Γ + SNR + PP + TES + LES + K + SK (无DDMA)	0.859	0.128	0.181
3	Γ + DDMA + PP + TES + LES + K + SK (无SNR)	0.817	0.146	0.204
4	Γ + DDMA + SNR + TES + LES + K + SK (无PP)	0.851	0.132	0.186
5	Γ + DDMA + SNR + PP + LES + K + SK (无TES)	0.871	0.122	0.174
6	Γ + DDMA + SNR + PP + TES + K + SK (无LES)	0.871	0.122	0.174
7	Γ + DDMA + SNR + PP + TES + LES + SK (无K)	0.866	0.124	0.177
8	Γ + DDMA + SNR + PP + TES + LES + K (无SK)	0.866	0.125	0.177
9	Γ + DDMA + SNR + PP + TES + LES + K + SK	0.875	0.120	0.171

下载: 导出CSV

参考文献(25)

[1]	Li Xiaoming, Sun Yan, Zhang Qiang. Extraction of sea ice cover by sentinel-1 SAR based on support vector machine with unsupervised generation of training datas[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(4): 3040−3053. doi: 10.1109/TGRS.2020.3007789
[2]	Chen Jinlei, Kang Shichang, You Qinglong, et al. Projected changes in sea ice and the navigability of the Arctic Passages under global warming of 2 ℃ and 3 ℃[J]. Anthropocene, 2022, 40: 100349. doi: 10.1016/j.ancene.2022.100349
[3]	谢涛, 赵立. 海冰密集度卫星遥感反演研究进展[J]. 海洋科学进展, 2022, 40(3): 351−366. doi: 10.12362/j.issn.1671-6647.20220209001 Xie Tao, Zhao Li. Advances in sea ice concentration retrieval based on satellite remote sensing[J]. Advances in Marine Science, 2022, 40(3): 351−366. doi: 10.12362/j.issn.1671-6647.20220209001
[4]	Zavorotny V U, Gleason S, Cardellach E, et al. Tutorial on remote sensing using GNSS bistatic radar of opportunity[J]. IEEE Geoscience and Remote Sensing Magazine, 2014, 2(4): 8−45. doi: 10.1109/MGRS.2014.2374220
[5]	Yan Qingyun, Huang Weimin, Moloney C. Neural networks based sea ice detection and concentration retrieval from GNSS-R delay-Doppler maps[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(8): 3789−3798. doi: 10.1109/JSTARS.2017.2689009
[6]	Yan Qingyun, Huang Weimin. Sea ice sensing from GNSS-R data using convolutional neural networks[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(10): 1510−1514. doi: 10.1109/LGRS.2018.2852143
[7]	Yan Qingyun, Huang Weimin. Sea ice concentration estimation from TechDemoSat-1 data using support vector regression[C]//2019 IEEE Radar Conference (RadarConf). Boston: IEEE, 2019: 1-6.
[8]	Llaveria D, Munoz-Martin J F, Herbert C, et al. Sea ice concentration and sea ice extent mapping with l-band microwave radiometry and GNSS-R data from the FFSCat mission using neural networks[J]. Remote Sensing, 2021, 13(6): 1139. doi: 10.3390/rs13061139
[9]	Zhu Yongchao, Tao T, Zou J, et al. Spaceborne GNSS reflectometry for retrieving sea ice concentration using TDS-1 data[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(4): 612−616. doi: 10.1109/LGRS.2020.2982959
[10]	Guo Wenfei, Du Hao, Guo Chi, et al. Information fusion for GNSS-R wind speed retrieval using statistically modified convolutional neural network[J]. Remote Sensing of Environment, 2022, 272: 112934. doi: 10.1016/j.rse.2022.112934
[11]	Liu Hongsu, Jin Shuanggen, Yan Qingyun. Evaluation of the ocean surface wind speed change following the super typhoon from space-borne GNSS-reflectometry[J]. Remote Sensing, 2020, 12(12): 2034. doi: 10.3390/rs12122034
[12]	Yan Qingyun, Huang Weimin. Spaceborne GNSS-R sea ice detection using delay-Doppler maps: first results from the U. K. TechDemoSat-1 mission[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 9(10): 4795−4801. doi: 10.1109/JSTARS.2016.2582690
[13]	Cartwright J, Banks C J, Srokosz M. Sea ice detection using GNSS-R data from TechDemoSat-1[J]. Journal of Geophysical Research: Oceans, 2019, 124(8): 5801−5810. doi: 10.1029/2019JC015327
[14]	Rodriguez-Alvarez N, Holt B, Jaruwatanadilok S, et al. An Arctic sea ice multi-step classification based on GNSS-R data from the TDS-1 mission[J]. Remote Sensing of Environment, 2019, 230: 111202. doi: 10.1016/j.rse.2019.05.021
[15]	邵连军, 胡磊, 李冰, 等. 基于CART决策树的星载GNSS-R海冰检测方法[J]. 海洋测绘, 2021, 41(1): 70−74. doi: 10.3969/j.issn.1671-3044.2021.01.015 Shao Lianjun, Hu Lei, Li Bing, et al. Sea ice detection using spaceborne GNSS-R data by CART decision tree[J]. Hydrographic Surveying and Charting, 2021, 41(1): 70−74. doi: 10.3969/j.issn.1671-3044.2021.01.015
[16]	Jales P, Unwin M. MERRByS product manual: GNSS reflectometry on TDS-1 with the SGR-ReSI[R]. Guildford: Surrey Satellite Technol. Ld., 2015.
[17]	Yao Ling, Lu Jiaying, Xia Xiaolin, et al. Evaluation of the ERA5 sea surface temperature around the Pacific and the Atlantic[J]. IEEE Access, 2021, 9: 12067−12073. doi: 10.1109/ACCESS.2021.3051642
[18]	Zhu Yongchao, Yu Kegen, Zou Jingui, et al. Sea ice detection based on differential delay-Doppler maps from UK TechDemoSat-1[J]. Sensors, 2017, 17(7): 1614. doi: 10.3390/s17071614
[19]	Santi E, Clarizia M P, Comite D, et al. Detecting fire disturbances in forests by using GNSS reflectometry and machine learning: a case study in Angola[J]. Remote Sensing of Environment, 2022, 270: 112878. doi: 10.1016/j.rse.2021.112878
[20]	Clarizia M P, Ruf C S, Jales P, et al. Spaceborne GNSS-R minimum variance wind speed estimator[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(11): 6829−6843. doi: 10.1109/TGRS.2014.2303831
[21]	Zhu Yongchao, Tao Tingye, Yu Kegen, et al. Sensing sea ice based on Doppler spread analysis of spaceborne GNSS-R data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 217−226. doi: 10.1109/JSTARS.2019.2955175
[22]	Yin Cong, Xia Junming, Huang Feixiong, et al. Sea ice detection with FY3E GNOS II GNSS reflectometry[C]//2021 IEEE Specialist Meeting on Reflectometry using GNSS and other Signals of Opportunity (GNSS+R). Beijing: IEEE, 2021: 36−38.
[23]	Alonso-Arroyo A, Zavorotny V U, Camps A. Sea ice detection using U. K. TDS-1 GNSS-R data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(9): 4989−5001. doi: 10.1109/TGRS.2017.2699122
[24]	Ke Guolin, Meng Qi, Finley T, et al. LightGBM: a highly efficient gradient boosting decision tree[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc., 2017: 3149−3157.
[25]	Chen Jiajia, Shen Huanfeng, Li Xinghua, et al. Ground-level ozone estimation based on geo-intelligent machine learning by fusing in-situ observations, remote sensing data, and model simulation data[J]. International Journal of Applied Earth Observation and Geoinformation, 2022, 112: 102955. doi: 10.1016/j.jag.2022.102955