1
|
Han Y, Huang J, Ma Z, Zheng B, Wang J, Zhang Y. GBDT Method Integrating Feature-Enhancement and Active-Learning Strategies-Sea Ice Thickness Inversion in Beaufort Sea. SENSORS (BASEL, SWITZERLAND) 2024; 24:2836. [PMID: 38732944 PMCID: PMC11086177 DOI: 10.3390/s24092836] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/07/2024] [Revised: 04/26/2024] [Accepted: 04/26/2024] [Indexed: 05/13/2024]
Abstract
Sea ice, as an important component of the Earth's ecosystem, has a profound impact on global climate and human activities due to its thickness. Therefore, the inversion of sea ice thickness has important research significance. Due to environmental and equipment-related limitations, the number of samples available for remote sensing inversion is currently insufficient. At high spatial resolutions, remote sensing data contain limited information and noise interference, which seriously affect the accuracy of sea ice thickness inversion. In response to the above issues, we conducted experiments using ice draft data from the Beaufort Sea and designed an improved GBDT method that integrates feature-enhancement and active-learning strategies (IFEAL-GBDT). In this method, the incident angle and time series are used to perform spatiotemporal correction of the data, reducing both temporal and spatial impacts. Meanwhile, based on the original polarization information, effective multi-attribute features are generated to expand the information content and improve the separability of sea ice with different thicknesses. Taking into account the growth cycle and age of sea ice, attributes were added for month and seawater temperature. In addition, we studied an active learning strategy based on the maximum standard deviation to select more informative and representative samples and improve the model's generalization ability. The improved GBDT model was used for training and prediction, offering advantages in dealing with nonlinear, high-dimensional data, and data noise problems, further expanding the effectiveness of feature-enhancement and active-learning strategies. Compared with other methods, the method proposed in this paper achieves the best inversion accuracy, with an average absolute error of 8 cm and a root mean square error of 13.7 cm for IFEAL-GBDT and a correlation coefficient of 0.912. This research proves the effectiveness of our method, which is suitable for the high-precision inversion of sea ice thickness determined using Sentinel-1 data.
Collapse
Affiliation(s)
| | | | - Zhenling Ma
- Shanghai Marine Intelligent Information and Navigation Remote Sensing Engineering Technology Research Center, Key Laboratory of Fisheries Information, Ministry of Agriculture, College of Information, Shanghai Ocean University, Shanghai 201306, China; (Y.H.); (J.H.); (B.Z.); (J.W.); (Y.Z.)
| | | | | | | |
Collapse
|
2
|
Hillebrand FL, Freitas MWDDE, Bremer UF, Abrantes TC, Simões JC, Mendes Júnior CW, Schardong F, Arigony-Neto J. Concentration and thickness of sea ice in the Weddell Sea from SSM/I passive microwave radiometer data. AN ACAD BRAS CIENC 2023; 95:e20230342. [PMID: 37937658 DOI: 10.1590/0001-3765202320230342] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/26/2023] [Accepted: 08/27/2023] [Indexed: 11/09/2023] Open
Abstract
This study evaluated feasibility statistically and analyzed, during the freezing period, the relationship between brightness temperature (Tb) data of the 37V polarisation and the GR3719 (Gradient Ratio 37V and 19V) obtained by Special Sensor Microwave/Imager from F11 and F13 satellites with sea ice thickness (SIT) data obtained in the Weddell Sea through Antarctic Sea Ice Processes and Climate program. The multiple linear regression (MLR) was applied at 1,520 points, with 70% of these points being randomly separated to generate the MLR and 30% to carry out the validation. To perform the temporal mapping, the MLR was applied only to pixels with sea ice concentration (SIC) ≥ 90%, obtained through the fraction image calculated from the spectral linear mixing model (SLMM) using the Tb in the channels and polarizations 19H, 19V and 37V. The results of the SLMM validation process for estimating the SIC were σ = 10.5%, RMSE = 11.0%, and bias = -2.8%, and the SIT based on the MLR, the results were R² = 0.57, RMSE = 0.268 m, and bias = 0.103 m. In the SIT mapping, we highlight the trend of thickness reduction on the east coast of the Antarctic Peninsula during the period 1992-2009.
Collapse
Affiliation(s)
- Fernando Luis Hillebrand
- Instituto Federal de Educacão, Ciência e Tecnologia do Rio Grande do Sul/IFRS, Rodovia RS-239, Km 68, 3505, 95700-000 Rolante, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Centro Polar e Climático, Av. Bento Gonçalves, 9500, Prédio 43136, Salas 208 e 210, 91501-970 Porto Alegre, RS, Brazil
| | - Marcos W D DE Freitas
- Universidade Federal do Rio Grande do Sul/UFRGS, Centro Polar e Climático, Av. Bento Gonçalves, 9500, Prédio 43136, Salas 208 e 210, 91501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Programa de Pós-Graduação em Sensoriamento Remoto, Av. Bento Gonçalves, 9500, Prédio 44202, Setor 5, 90501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Instituto de Geociências, Av. Bento Gonçalves, 9500, 90501-970 Porto Alegre, RS, Brazil
| | - Ulisses F Bremer
- Universidade Federal do Rio Grande do Sul/UFRGS, Centro Polar e Climático, Av. Bento Gonçalves, 9500, Prédio 43136, Salas 208 e 210, 91501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Programa de Pós-Graduação em Sensoriamento Remoto, Av. Bento Gonçalves, 9500, Prédio 44202, Setor 5, 90501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Instituto de Geociências, Av. Bento Gonçalves, 9500, 90501-970 Porto Alegre, RS, Brazil
| | - Tales C Abrantes
- Universidade Federal do Rio Grande do Sul/UFRGS, Programa de Pós-Graduação em Sensoriamento Remoto, Av. Bento Gonçalves, 9500, Prédio 44202, Setor 5, 90501-970 Porto Alegre, RS, Brazil
| | - Jefferson C Simões
- Universidade Federal do Rio Grande do Sul/UFRGS, Centro Polar e Climático, Av. Bento Gonçalves, 9500, Prédio 43136, Salas 208 e 210, 91501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Instituto de Geociências, Av. Bento Gonçalves, 9500, 90501-970 Porto Alegre, RS, Brazil
| | - Cláudio W Mendes Júnior
- Universidade Federal do Rio Grande do Sul/UFRGS, Centro Polar e Climático, Av. Bento Gonçalves, 9500, Prédio 43136, Salas 208 e 210, 91501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Programa de Pós-Graduação em Sensoriamento Remoto, Av. Bento Gonçalves, 9500, Prédio 44202, Setor 5, 90501-970 Porto Alegre, RS, Brazil
- Universidade Federal do Rio Grande do Sul/UFRGS, Instituto de Geociências, Av. Bento Gonçalves, 9500, 90501-970 Porto Alegre, RS, Brazil
| | - Frederico Schardong
- Instituto Federal de Educacão, Ciência e Tecnologia do Rio Grande do Sul/IFRS, Rodovia RS-239, Km 68, 3505, 95700-000 Rolante, RS, Brazil
| | - Jorge Arigony-Neto
- Universidade Federal do Rio Grande/FURG, Instituto de Oceanografia, Av. Itália, s/n, Km 8, 96201-900 Rio Grande, RS, Brazil
| |
Collapse
|
3
|
Abstract
In this study, we retrieve an Arctic sea ice lead fraction from AMSR2 passive microwave data in winter from 2012 to 2020 based on an algorithm developed for AMSR-E data. The derived AMSR2 sea ice lead fraction is validated against MODIS images. The results show that the derived AMSR2 sea ice lead detects approximately 50% of the ice leads shown in the MODIS images, which is close to the amount of sea ice lead detected from the AMSR-E data from 2002 to 2011. Utilizing the retrievals from both the AMSR-E and AMSR2, our analysis shows no significant trend, but moderate interannual variation exists for the ice lead fraction in the Arctic basin scale over the past two decades. The maximum width and total length of sea ice lead show a significant decreasing trend for the whole Arctic, but the mean width does not exhibit a significant change over the studied period. In the Beaufort Sea the lead fraction varies from 2.06% to 12.35%, with a mean value of 5.72%. In the Greenland Sea the mean lead fraction over the studied period is 5.77%, and there is a significant increase in the lead fraction, with a rate of 0.13% per year. The maximum width in the Greenland Sea is substantially higher than that of other regions, and the mean width increases significantly.
Collapse
|