Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry
Satellite and airborne observations of surface elevation are critical in understanding climatic and glaciological processes and quantifying their impact on changes in ice masses and sea level contribution. With the growing number of dedicated airborne campaigns and experimental and operational satel...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/24/6210 |
| _version_ | 1797455420444377088 |
|---|---|
| author | Alex Horton Martin Ewart Noel Gourmelen Xavier Fettweis Amos Storkey |
| author_facet | Alex Horton Martin Ewart Noel Gourmelen Xavier Fettweis Amos Storkey |
| author_sort | Alex Horton |
| collection | DOAJ |
| description | Satellite and airborne observations of surface elevation are critical in understanding climatic and glaciological processes and quantifying their impact on changes in ice masses and sea level contribution. With the growing number of dedicated airborne campaigns and experimental and operational satellite missions, the science community has access to unprecedented and ever-increasing data. Combining elevation datasets allows potentially greater spatial-temporal coverage and improved accuracy; however, combining data from different sensor types and acquisition modes is difficult by differences in intrinsic sensor properties and processing methods. This study focuses on the combination of elevation measurements derived from ICESat-2 and Operation IceBridge LIDAR instruments and from CryoSat-2’s novel interferometric radar altimeter over Greenland. We develop a deep neural network based on sub-waveform information from CryoSat-2, elevation differences between radar and LIDAR, and additional inputs representing local geophysical information. A time series of maps are created showing observed LIDAR-radar differences and neural network model predictions. Mean LIDAR vs. interferometric radar adjustments and the broad spatial and temporal trends thereof are recreated by the neural network. The neural network also predicts radar-LIDAR differences with respect to waveform parameters better than a simple linear model; however, point level adjustments and the magnitudes of the spatial and temporal trends are underestimated. |
| first_indexed | 2024-03-09T15:54:08Z |
| format | Article |
| id | doaj.art-10cfa237cff64197bc3290c94f4ca1df |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T15:54:08Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-10cfa237cff64197bc3290c94f4ca1df2023-11-24T17:45:57ZengMDPI AGRemote Sensing2072-42922022-12-011424621010.3390/rs14246210Using Deep Learning to Model Elevation Differences between Radar and Laser AltimetryAlex Horton0Martin Ewart1Noel Gourmelen2Xavier Fettweis3Amos Storkey4Earthwave Ltd., Edinburgh EH3 9DR, UKEarthwave Ltd., Edinburgh EH3 9DR, UKSchool of Geosciences, University of Edinburgh, Edinburgh EH8 9YL, UKSPHERES Research Unit, Department of Geography, University of Liège, 4000 Liège, BelgiumSchool of Informatics, University of Edinburgh, Edinburgh EH8 9YL, UKSatellite and airborne observations of surface elevation are critical in understanding climatic and glaciological processes and quantifying their impact on changes in ice masses and sea level contribution. With the growing number of dedicated airborne campaigns and experimental and operational satellite missions, the science community has access to unprecedented and ever-increasing data. Combining elevation datasets allows potentially greater spatial-temporal coverage and improved accuracy; however, combining data from different sensor types and acquisition modes is difficult by differences in intrinsic sensor properties and processing methods. This study focuses on the combination of elevation measurements derived from ICESat-2 and Operation IceBridge LIDAR instruments and from CryoSat-2’s novel interferometric radar altimeter over Greenland. We develop a deep neural network based on sub-waveform information from CryoSat-2, elevation differences between radar and LIDAR, and additional inputs representing local geophysical information. A time series of maps are created showing observed LIDAR-radar differences and neural network model predictions. Mean LIDAR vs. interferometric radar adjustments and the broad spatial and temporal trends thereof are recreated by the neural network. The neural network also predicts radar-LIDAR differences with respect to waveform parameters better than a simple linear model; however, point level adjustments and the magnitudes of the spatial and temporal trends are underestimated.https://www.mdpi.com/2072-4292/14/24/6210SARIninterferometryCryoSatswathICESat-2IceBridge |
| spellingShingle | Alex Horton Martin Ewart Noel Gourmelen Xavier Fettweis Amos Storkey Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry Remote Sensing SARIn interferometry CryoSat swath ICESat-2 IceBridge |
| title | Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry |
| title_full | Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry |
| title_fullStr | Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry |
| title_full_unstemmed | Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry |
| title_short | Using Deep Learning to Model Elevation Differences between Radar and Laser Altimetry |
| title_sort | using deep learning to model elevation differences between radar and laser altimetry |
| topic | SARIn interferometry CryoSat swath ICESat-2 IceBridge |
| url | https://www.mdpi.com/2072-4292/14/24/6210 |
| work_keys_str_mv | AT alexhorton usingdeeplearningtomodelelevationdifferencesbetweenradarandlaseraltimetry AT martinewart usingdeeplearningtomodelelevationdifferencesbetweenradarandlaseraltimetry AT noelgourmelen usingdeeplearningtomodelelevationdifferencesbetweenradarandlaseraltimetry AT xavierfettweis usingdeeplearningtomodelelevationdifferencesbetweenradarandlaseraltimetry AT amosstorkey usingdeeplearningtomodelelevationdifferencesbetweenradarandlaseraltimetry |