Sea ice thickness in the Beaufort Sea and comparisons with C-band SAR imagery
Abstract
The Beaufort Sea is a region of significant interest for both climate science and marine operations, as summer sea ice extent and ice age have shown strong declines in recent years, while interest in oil and gas activities have increased dramatically. As a result, there is high demand for sea ice information products in this region. In particular, routine observations of ice thickness are desired at regional and local scales for parameterization and validation of seasonal and long-term ice forecasts, as well as ensuring safe marine operations. We present ice thickness measurements acquired in April 2009-2016 with an airborne electromagnetic induction system (AEM). The AEM data are compared to synthetic aperture radar (SAR) imagery acquired by RADARSAT-2 (RS2) in 2015 and 2016 to determine if ice thickness can be inverted from C-band SAR data. The RS2 data include ScanSAR dual-pol (HH/HV) and high-resolution quad-pol images. The AEM measurements indicate that the thickest multi-year ice (MYI) often occurs in a narrow band at the southern edge of the MYI zone, and that there is high interannual variability in modal MYI thickness. The southern Beaufort Sea is dominated by first-year ice (FYI) that is consistently 2 m thick. Significant areas of thin FYI were present in 2016 due to strong divergent ice motion and westward advection in February-April. Extreme ice features (> 6 m thick for 100+ m along flight track) were observed in all study years in both MYI and FYI regions. The AEM data were averaged in 0.5 to 5 km intervals and compared to SAR backscatter and polarimetric parameters. Correlations between ice thickness and SAR backscatter and polarimetric parameters were generally weak to moderate, and inconsistent between study years. Work is ongoing to compare AEM and SAR data at the floe scale instead of at fixed distance intervals. Depolarization parameters, which are strongly related to volume scattering, show potential for identifying thick deformed ice features. Efforts to reliably separate these hazardous deformed ice features from small-scale roughness features associated with thinner ice types (e.g. frost flowers, pancake ice and brash ice) are ongoing. It is recommended that future work include L-band SAR data to aid in the discrimination of small- and large-scale roughness features.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFM.C33B0807C
- Keywords:
-
- 0758 Remote sensing;
- CRYOSPHERE