Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data
Bed topography and roughness play important roles in numerous ice-sheet analyses. Although the coverage of ice-penetrating radar measurements has vastly increased over recent decades, significant data gaps remain in certain areas of subglacial topography and need interpolation. However, the bed topo...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/5/1359 |
| _version_ | 1797614480415260672 |
|---|---|
| author | Yiheng Cai Fuxing Wan Shinan Lang Xiangbin Cui Zijun Yao |
| author_facet | Yiheng Cai Fuxing Wan Shinan Lang Xiangbin Cui Zijun Yao |
| author_sort | Yiheng Cai |
| collection | DOAJ |
| description | Bed topography and roughness play important roles in numerous ice-sheet analyses. Although the coverage of ice-penetrating radar measurements has vastly increased over recent decades, significant data gaps remain in certain areas of subglacial topography and need interpolation. However, the bed topography generated by interpolation such as kriging and mass conservation is generally smooth at small scales, lacking topographic features important for sub-kilometer roughness. DeepBedMap, a deep learning method combined with multiple surface observation inputs, can generate high-resolution (250 m) bed topography with realistic bed roughness but produces some unrealistic artifacts and higher bed elevation values in certain regions, which could bias ice-sheet models. To address these issues, we present MB_DeepBedMap, a multi-branch deep learning method to generate more realistic bed topography. The model improves upon DeepBedMap by separating inputs into two groups using a multi-branch network structure according to their characteristics, rather than fusing all inputs at an early stage, to reduce artifacts in the generated topography caused by earlier fusion of inputs. A direct upsampling branch preserves large-scale subglacial landforms while generating high-resolution bed topography. We use MB_DeepBedMap to generate a high-resolution (250 m) bed elevation grid product of Antarctica, MB_DeepBedMap_DEM, which can be used in high-resolution ice-sheet modeling studies. Moreover, we test the performance of MB_DeepBedMap model in Thwaites Glacier, Gamburtsev Subglacial Mountains, and several other regions, by comparing the qualitative topographic features and quantitative errors of MB_DeepBedMap, BEDMAP2, BedMachine Antarctica, and DeepBedMap. The results show that MB_DeepBedMap can provide more realistic small-scale topographic features and roughness compared to BEDMAP2, BedMachine Antarctica, and DeepBedMap. |
| first_indexed | 2024-03-11T07:12:02Z |
| format | Article |
| id | doaj.art-eeeb4935b2d74c17b7c2c66e272980b6 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T07:12:02Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-eeeb4935b2d74c17b7c2c66e272980b62023-11-17T08:32:00ZengMDPI AGRemote Sensing2072-42922023-02-01155135910.3390/rs15051359Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing DataYiheng Cai0Fuxing Wan1Shinan Lang2Xiangbin Cui3Zijun Yao4School of Information and Communications Engineering, Beijing University of Technology, Beijing 100124, ChinaSchool of Information and Communications Engineering, Beijing University of Technology, Beijing 100124, ChinaSchool of Information and Communications Engineering, Beijing University of Technology, Beijing 100124, ChinaPolar Research Institute of China, Shanghai 200136, ChinaSchool of Information and Communications Engineering, Beijing University of Technology, Beijing 100124, ChinaBed topography and roughness play important roles in numerous ice-sheet analyses. Although the coverage of ice-penetrating radar measurements has vastly increased over recent decades, significant data gaps remain in certain areas of subglacial topography and need interpolation. However, the bed topography generated by interpolation such as kriging and mass conservation is generally smooth at small scales, lacking topographic features important for sub-kilometer roughness. DeepBedMap, a deep learning method combined with multiple surface observation inputs, can generate high-resolution (250 m) bed topography with realistic bed roughness but produces some unrealistic artifacts and higher bed elevation values in certain regions, which could bias ice-sheet models. To address these issues, we present MB_DeepBedMap, a multi-branch deep learning method to generate more realistic bed topography. The model improves upon DeepBedMap by separating inputs into two groups using a multi-branch network structure according to their characteristics, rather than fusing all inputs at an early stage, to reduce artifacts in the generated topography caused by earlier fusion of inputs. A direct upsampling branch preserves large-scale subglacial landforms while generating high-resolution bed topography. We use MB_DeepBedMap to generate a high-resolution (250 m) bed elevation grid product of Antarctica, MB_DeepBedMap_DEM, which can be used in high-resolution ice-sheet modeling studies. Moreover, we test the performance of MB_DeepBedMap model in Thwaites Glacier, Gamburtsev Subglacial Mountains, and several other regions, by comparing the qualitative topographic features and quantitative errors of MB_DeepBedMap, BEDMAP2, BedMachine Antarctica, and DeepBedMap. The results show that MB_DeepBedMap can provide more realistic small-scale topographic features and roughness compared to BEDMAP2, BedMachine Antarctica, and DeepBedMap.https://www.mdpi.com/2072-4292/15/5/1359subglacial topographysuper resolutiongenerative adversarial networkdeep learning |
| spellingShingle | Yiheng Cai Fuxing Wan Shinan Lang Xiangbin Cui Zijun Yao Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data Remote Sensing subglacial topography super resolution generative adversarial network deep learning |
| title | Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data |
| title_full | Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data |
| title_fullStr | Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data |
| title_full_unstemmed | Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data |
| title_short | Multi-Branch Deep Neural Network for Bed Topography of Antarctica Super-Resolution: Reasonable Integration of Multiple Remote Sensing Data |
| title_sort | multi branch deep neural network for bed topography of antarctica super resolution reasonable integration of multiple remote sensing data |
| topic | subglacial topography super resolution generative adversarial network deep learning |
| url | https://www.mdpi.com/2072-4292/15/5/1359 |
| work_keys_str_mv | AT yihengcai multibranchdeepneuralnetworkforbedtopographyofantarcticasuperresolutionreasonableintegrationofmultipleremotesensingdata AT fuxingwan multibranchdeepneuralnetworkforbedtopographyofantarcticasuperresolutionreasonableintegrationofmultipleremotesensingdata AT shinanlang multibranchdeepneuralnetworkforbedtopographyofantarcticasuperresolutionreasonableintegrationofmultipleremotesensingdata AT xiangbincui multibranchdeepneuralnetworkforbedtopographyofantarcticasuperresolutionreasonableintegrationofmultipleremotesensingdata AT zijunyao multibranchdeepneuralnetworkforbedtopographyofantarcticasuperresolutionreasonableintegrationofmultipleremotesensingdata |