Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017
Temporal and spatial variation of sea ice type in the Arctic is an indicator of regional and global change. Arctic sea ice can be classified into two major categories: multiyear ice (MYI) and first-year ice. In this paper, classification method based on machine learning is established and applied to...
Published in: | IEEE Transactions on Geoscience and Remote Sensing |
---|---|
Main Authors: | , , , , , |
Language: | unknown |
Published: |
2021
|
Subjects: | |
Online Access: | http://www.osti.gov/servlets/purl/1503006 https://www.osti.gov/biblio/1503006 https://doi.org/10.1109/TGRS.2019.2898872 |
id |
ftosti:oai:osti.gov:1503006 |
---|---|
record_format |
openpolar |
spelling |
ftosti:oai:osti.gov:1503006 2023-07-30T04:00:09+02:00 Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 Zhang, Zhilun Yu, Yining Li, Xinqing Hui, Fengming Cheng, Xiao Chen, Zhuoqi 2021-07-29 application/pdf http://www.osti.gov/servlets/purl/1503006 https://www.osti.gov/biblio/1503006 https://doi.org/10.1109/TGRS.2019.2898872 unknown http://www.osti.gov/servlets/purl/1503006 https://www.osti.gov/biblio/1503006 https://doi.org/10.1109/TGRS.2019.2898872 doi:10.1109/TGRS.2019.2898872 58 GEOSCIENCES 2021 ftosti https://doi.org/10.1109/TGRS.2019.2898872 2023-07-11T09:32:12Z Temporal and spatial variation of sea ice type in the Arctic is an indicator of regional and global change. Arctic sea ice can be classified into two major categories: multiyear ice (MYI) and first-year ice. In this paper, classification method based on machine learning is established and applied to produce daily sea ice classification data set during the winter (November–April) from 2002 to 2017 using active microwave data from QuikSCAT and Advanced Scatterometer as well as passive microwave data from Advanced Microwave Scanning Radiometer for EOS, Special Sensor Microwave Imager/Sounder, and Advanced Microwave Scanning Radiometer 2 radiometer. First, the open water area is flagged out using brightness temperature (Tb) from the passive microwave sensor. Then, K-means algorithm is applied to identify the clusters of the two ice types in the Tb/backscatter parameter space and finally assign pixels to each class. Two optimization methods based on the movement of MYI and marginal ice zone are used to correct the misclassification of MYI. The results have shown a decrease of MYI in winter from 2002 to 2017, especially in 2008 and 2013 with a remarkable recovery in 2014. The classifications are consistent with results by visual interpretation from synthetic aperture radar images in the Canadian Arctic Archipelago with overall classification accuracy over 93%. Comparison with classifications from previous studies and products shows that our method could reflect more differences in MYI declining trend interannually and less anomalous fluctuations in certain years. Other/Unknown Material Arctic Archipelago Arctic Canadian Arctic Archipelago Sea ice SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) Arctic Canadian Arctic Archipelago IEEE Transactions on Geoscience and Remote Sensing 57 8 5319 5328 |
institution |
Open Polar |
collection |
SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) |
op_collection_id |
ftosti |
language |
unknown |
topic |
58 GEOSCIENCES |
spellingShingle |
58 GEOSCIENCES Zhang, Zhilun Yu, Yining Li, Xinqing Hui, Fengming Cheng, Xiao Chen, Zhuoqi Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
topic_facet |
58 GEOSCIENCES |
description |
Temporal and spatial variation of sea ice type in the Arctic is an indicator of regional and global change. Arctic sea ice can be classified into two major categories: multiyear ice (MYI) and first-year ice. In this paper, classification method based on machine learning is established and applied to produce daily sea ice classification data set during the winter (November–April) from 2002 to 2017 using active microwave data from QuikSCAT and Advanced Scatterometer as well as passive microwave data from Advanced Microwave Scanning Radiometer for EOS, Special Sensor Microwave Imager/Sounder, and Advanced Microwave Scanning Radiometer 2 radiometer. First, the open water area is flagged out using brightness temperature (Tb) from the passive microwave sensor. Then, K-means algorithm is applied to identify the clusters of the two ice types in the Tb/backscatter parameter space and finally assign pixels to each class. Two optimization methods based on the movement of MYI and marginal ice zone are used to correct the misclassification of MYI. The results have shown a decrease of MYI in winter from 2002 to 2017, especially in 2008 and 2013 with a remarkable recovery in 2014. The classifications are consistent with results by visual interpretation from synthetic aperture radar images in the Canadian Arctic Archipelago with overall classification accuracy over 93%. Comparison with classifications from previous studies and products shows that our method could reflect more differences in MYI declining trend interannually and less anomalous fluctuations in certain years. |
author |
Zhang, Zhilun Yu, Yining Li, Xinqing Hui, Fengming Cheng, Xiao Chen, Zhuoqi |
author_facet |
Zhang, Zhilun Yu, Yining Li, Xinqing Hui, Fengming Cheng, Xiao Chen, Zhuoqi |
author_sort |
Zhang, Zhilun |
title |
Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
title_short |
Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
title_full |
Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
title_fullStr |
Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
title_full_unstemmed |
Arctic Sea Ice Classification Using Microwave Scatterometer and Radiometer Data During 2002-2017 |
title_sort |
arctic sea ice classification using microwave scatterometer and radiometer data during 2002-2017 |
publishDate |
2021 |
url |
http://www.osti.gov/servlets/purl/1503006 https://www.osti.gov/biblio/1503006 https://doi.org/10.1109/TGRS.2019.2898872 |
geographic |
Arctic Canadian Arctic Archipelago |
geographic_facet |
Arctic Canadian Arctic Archipelago |
genre |
Arctic Archipelago Arctic Canadian Arctic Archipelago Sea ice |
genre_facet |
Arctic Archipelago Arctic Canadian Arctic Archipelago Sea ice |
op_relation |
http://www.osti.gov/servlets/purl/1503006 https://www.osti.gov/biblio/1503006 https://doi.org/10.1109/TGRS.2019.2898872 doi:10.1109/TGRS.2019.2898872 |
op_doi |
https://doi.org/10.1109/TGRS.2019.2898872 |
container_title |
IEEE Transactions on Geoscience and Remote Sensing |
container_volume |
57 |
container_issue |
8 |
container_start_page |
5319 |
op_container_end_page |
5328 |
_version_ |
1772810737688772608 |