Bayesian sea ice detection with the advanced scatterometer ASCAT
This paper details the construction of a Bayesian sea ice detection algorithm for the C-band Advanced Scatterometer ASCAT onboard MetOp based on probabilistic distances to ocean wind and sea ice geophysical model functions. The performance of the algorithm is validated against coincident active and...
Published in: | IEEE Transactions on Geoscience and Remote Sensing |
---|---|
Other Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Institute of Electrical and Electronics Engineers
|
Subjects: | |
Online Access: | https://doi.org/10.1109/TGRS.2011.2182356 http://n2t.net/ark:/85065/d7p26zs6 |
Summary: | This paper details the construction of a Bayesian sea ice detection algorithm for the C-band Advanced Scatterometer ASCAT onboard MetOp based on probabilistic distances to ocean wind and sea ice geophysical model functions. The performance of the algorithm is validated against coincident active and passive microwave sea ice extents on a global scale across the seasons. The comparison between the ASCAT, QuikSCAT, and AMSR-E records during 2008 is satisfactory during the winter seasons, but reveals systematic biases between active and passive microwave methods during the summer months. These differences arise from their different sensitivities to mixed sea ice and open water conditions, scatterometers being more inclusive regarding the detection of lower concentration and summer ice. The sea ice normalized backscatter observed at C-band shows some loss of contrast between thin and thick ice types relative to the Ku-band QuikSCAT, but offers a better sensitivity to prominent surface features, such as fragmentation and rafting of marginal sea ice. |
---|