Arctic Sea Ice Characterization for Navigational Assistance using Space-borne Fully Polarimetric SAR
Satellite images are an essential parameter for sea ice products such as the daily sea ice concentration maps issued by the Norwegian Ice Service. In this regard, Synthetic Aperture Radar (SAR) is especially valuable given its all-weather capabilities and that it does not require sunlight. Recently...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | , |
| Format: | Conference Object |
| Language: | unknown |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://elib.dlr.de/116306/ https://www.arcticfrontiers.com/program/session/?id=S055 |
| Summary: | Satellite images are an essential parameter for sea ice products such as the daily sea ice concentration maps issued by the Norwegian Ice Service. In this regard, Synthetic Aperture Radar (SAR) is especially valuable given its all-weather capabilities and that it does not require sunlight. Recently launched missions such as Sentinel-1 acquire data in dual-polarimetric mode, something that has been proven to be more useful than conventional single polarization SAR in terms of characterizing different sea ice types. Within this study we further investigate the usefulness of fully polarimetric SAR. Here we employ an automatic sea ice classification algorithm developed for Near Real Time (NRT) services on two sets of spatially and temporally near coincident fully polarimetric acquisitions from the ALOS-2, Radarsat-2 and TerraSAR-X/TanDEM-X satellites acquired during the N-ICE2015 sea ice drift study. Overlapping coincident sea ice freeboard measurements from Airborne Laser Scanner (ALS) data are used to validate the classification results. We analyzed the usefulness for the classification results of 18 different polarimetric parameters. In order to deliver sea ice products in NRT efficient computation is very important and by reducing redundant or overlapping parameters we can speed up the delivery time. Among the most useful features for classification were the geometric intensity, the scattering diversity and the surface scattering fraction. We also found that we can halve the number of parameters from 18 to 9. In our study, the ALS data show that 100 % of the open water is separated from the surrounding sea ice and that the sea ice classes have at least 96.9 % accuracy. This analysis reveals analogous results for both X-band and C-band frequencies and slightly different for L-band. In particular, the overlapping image portions exhibit a reasonable congruence of detected sea ice when compared with high resolution airborne measurements |
|---|