Answers are grounded in this paper and cite their sources.
Combining backscatter modeling with observations from Sentinel-1 (S1) wide-swath imagery, we derive a threshold separating sea ice and open water based on their incidence angle (IA) slopes for HH-polarized backscatter intensity in winter conditions. Backscatter from C-band synthetic aperture radar (SAR) over open water is here modeled through a geophysical model function for wind speeds up to 25 m/s. Averaged linear slopes are estimated from the geophysical modeled backscatter across the full scene as well as in local IA bins to account for curvature. IA slopes from different sea ice types across multiple Arctic study sights are found in literature in addition to regression analysis on SAR backscatter from selected S1 scenes. Results indicate that −0.40 dB/1 ^∘ is a suitable threshold distinguishing ice from open water with uniform wind conditions. To demonstrate the threshold, and the potential for automatically assigning physics based class labels (sea ice and open water), we apply it to results from unsupervised clustering showcasing high spatial resolution ice–water maps. Challenges related to clustering and local wind gradients are further explored and discussed.
No takes yet. Share an insight, caveat, or question.
Östersund et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: