Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data
Abstract Microwave radiometry at L‐band is sensitive to sea ice thickness (SIT) up to ∼ 60 cm. Current methods to infer SIT depend on ice‐physical properties and data provided by the ESA’s Soil Moisture and Ocean Salinity (SMOS) mission. However, retrieval accuracy is limited due to seasonally and r...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-03-01
|
| Series: | Geophysical Research Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2020GL091285 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850205166808072192 |
|---|---|
| author | Christoph Herbert Adriano Camps Florian Wellmann Mercedes Vall‐llossera |
| author_facet | Christoph Herbert Adriano Camps Florian Wellmann Mercedes Vall‐llossera |
| author_sort | Christoph Herbert |
| collection | DOAJ |
| description | Abstract Microwave radiometry at L‐band is sensitive to sea ice thickness (SIT) up to ∼ 60 cm. Current methods to infer SIT depend on ice‐physical properties and data provided by the ESA’s Soil Moisture and Ocean Salinity (SMOS) mission. However, retrieval accuracy is limited due to seasonally and regionally variable surface conditions during the formation and melting of sea ice. In this work, Arctic sea ice is segmented using a Bayesian unsupervised learning algorithm aiming to recognize spatial patterns by harnessing multi‐incidence angle brightness temperature observations. The approach considers both statistical characteristics and spatial correlations of the observations. The temporal stability and separability of classes are analyzed to distinguish ambiguous from well‐determined regions. Model uncertainty is quantified from class membership probabilities using information entropy. The presented approach opens up a new scope to improve current SIT retrieval algorithms, and can be particularly beneficial to investigate merged satellite products. |
| format | Article |
| id | doaj-art-e7264c3796404e5aacc41f81f8acb128 |
| institution | OA Journals |
| issn | 0094-8276 1944-8007 |
| language | English |
| publishDate | 2021-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | Geophysical Research Letters |
| spelling | doaj-art-e7264c3796404e5aacc41f81f8acb1282025-08-20T02:11:09ZengWileyGeophysical Research Letters0094-82761944-80072021-03-01486n/an/a10.1029/2020GL091285Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS DataChristoph Herbert0Adriano Camps1Florian Wellmann2Mercedes Vall‐llossera3CommSensLab Universitat Politècnica de Catalunya (UPC) and Institut d’Estudis Espacials de Catalunya (IEEC/CTE‐UPC) Barcelona SpainCommSensLab Universitat Politècnica de Catalunya (UPC) and Institut d’Estudis Espacials de Catalunya (IEEC/CTE‐UPC) Barcelona SpainInstitute for Computational Geoscience and Reservoir Engineering RWTH Aachen University Aachen GermanyCommSensLab Universitat Politècnica de Catalunya (UPC) and Institut d’Estudis Espacials de Catalunya (IEEC/CTE‐UPC) Barcelona SpainAbstract Microwave radiometry at L‐band is sensitive to sea ice thickness (SIT) up to ∼ 60 cm. Current methods to infer SIT depend on ice‐physical properties and data provided by the ESA’s Soil Moisture and Ocean Salinity (SMOS) mission. However, retrieval accuracy is limited due to seasonally and regionally variable surface conditions during the formation and melting of sea ice. In this work, Arctic sea ice is segmented using a Bayesian unsupervised learning algorithm aiming to recognize spatial patterns by harnessing multi‐incidence angle brightness temperature observations. The approach considers both statistical characteristics and spatial correlations of the observations. The temporal stability and separability of classes are analyzed to distinguish ambiguous from well‐determined regions. Model uncertainty is quantified from class membership probabilities using information entropy. The presented approach opens up a new scope to improve current SIT retrieval algorithms, and can be particularly beneficial to investigate merged satellite products.https://doi.org/10.1029/2020GL091285Bayesian inferencesea iceSMOSunsupervised learning |
| spellingShingle | Christoph Herbert Adriano Camps Florian Wellmann Mercedes Vall‐llossera Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data Geophysical Research Letters Bayesian inference sea ice SMOS unsupervised learning |
| title | Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data |
| title_full | Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data |
| title_fullStr | Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data |
| title_full_unstemmed | Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data |
| title_short | Bayesian Unsupervised Machine Learning Approach to Segment Arctic Sea Ice Using SMOS Data |
| title_sort | bayesian unsupervised machine learning approach to segment arctic sea ice using smos data |
| topic | Bayesian inference sea ice SMOS unsupervised learning |
| url | https://doi.org/10.1029/2020GL091285 |
| work_keys_str_mv | AT christophherbert bayesianunsupervisedmachinelearningapproachtosegmentarcticseaiceusingsmosdata AT adrianocamps bayesianunsupervisedmachinelearningapproachtosegmentarcticseaiceusingsmosdata AT florianwellmann bayesianunsupervisedmachinelearningapproachtosegmentarcticseaiceusingsmosdata AT mercedesvallllossera bayesianunsupervisedmachinelearningapproachtosegmentarcticseaiceusingsmosdata |