Arctic sea ice is segmented from SMOS brightness temperature observations based on a Bayesian unsupervised learning approach. The obtained classes can be analyzed in terms of their temporal stability and separability. The resulting spatial patterns are then related to different sea ice types and thickness
Segmentation results for the period from September 1 to December 31, 2016, including late summer melt and early freeze up.
For further information, please contact: Christoph Herbert :herbert@tsc.upc.edu