A Study of Polarimetric Parameters Used for Sea Ice Classification

The Arctic region of the Earth are largely covered by sea ice, which is important for the marine ecosystems locally in the region, but it also has an impact on the climate systems. Recent climate changes means that the sea ice is melting and the coverage is shrinking which can have a global impact....

Full description

Bibliographic Details
Main Author: Kinderås, Ida G.
Format: Master Thesis
Language:English
Published: UiT The Arctic University of Norway 2021
Subjects:
Online Access:https://hdl.handle.net/10037/25092
id ftunivtroemsoe:oai:munin.uit.no:10037/25092
record_format openpolar
spelling ftunivtroemsoe:oai:munin.uit.no:10037/25092 2023-05-15T14:56:23+02:00 A Study of Polarimetric Parameters Used for Sea Ice Classification Kinderås, Ida G. 2021-12-15 https://hdl.handle.net/10037/25092 eng eng UiT The Arctic University of Norway UiT Norges arktiske universitet https://hdl.handle.net/10037/25092 openAccess Copyright 2021 The Author(s) VDP::Teknologi: 500::Miljøteknologi: 610 VDP::Technology: 500::Environmental engineering: 610 EOM-3901 Master thesis Mastergradsoppgave 2021 ftunivtroemsoe 2022-05-11T22:58:43Z The Arctic region of the Earth are largely covered by sea ice, which is important for the marine ecosystems locally in the region, but it also has an impact on the climate systems. Recent climate changes means that the sea ice is melting and the coverage is shrinking which can have a global impact. Monitoring these sea ice changes are important to get an understanding of this impact. There are multiple options for how to carry out this monitoring, but the Arctic is a cold, dark and remote region, so a good option is using satellite images. A special advantage of using SAR (Synthetic Aperture Radar) satellites is that they are insensitive to weather conditions and darkness, which can be a challenge in the Arctic regions, especially during the winter months. Today operational ice services produces sea ice charts using such SAR images but they primarily reply on dual-polarimetric data. These images covers large areas due to the large swath for these modes and are idea for a Pan-Arctic coverage. However, they are impacted by e.g., effects of varying incidence angle throughout the images. This thesis will study the possibilities that lies within radar polarimetry, a field that utilized fully polarimetric data to extract even more information regarding the target on the surface, in this case sea ice. Moreover, these images enabled extraction of scattering type information and through this added information improve the sea ice type classification maps. The images studied here are two sets of fully polarimetric RADARSAT-2 data that covers the same sea ice but with varying incidence angles. Four different polarimetric features; the co-polarization ratio, the M-parameter, degree of polarization as well as the polarimetric coherence, were investigated for their separability measures between different sea ice types as well as how affected they are by incidence angle variations. Out of these four the co-polarization ratio was found to be useful to separate the thinner sea ice types as well as open water from the surrounding thicker sea ice. Both the M-parameter and the degree of polarization showed promise in separating deformed sea ice from the other ice types at higher incidence angles. Though limited consistent separability between different ice types were observed for the polarimetric coherence. The polarimetric coherence was found to be independent on the incidence angle, though the other three parameters showed an incidence angle dependency. Master Thesis Arctic Sea ice University of Tromsø: Munin Open Research Archive Arctic
institution Open Polar
collection University of Tromsø: Munin Open Research Archive
op_collection_id ftunivtroemsoe
language English
topic VDP::Teknologi: 500::Miljøteknologi: 610
VDP::Technology: 500::Environmental engineering: 610
EOM-3901
spellingShingle VDP::Teknologi: 500::Miljøteknologi: 610
VDP::Technology: 500::Environmental engineering: 610
EOM-3901
Kinderås, Ida G.
A Study of Polarimetric Parameters Used for Sea Ice Classification
topic_facet VDP::Teknologi: 500::Miljøteknologi: 610
VDP::Technology: 500::Environmental engineering: 610
EOM-3901
description The Arctic region of the Earth are largely covered by sea ice, which is important for the marine ecosystems locally in the region, but it also has an impact on the climate systems. Recent climate changes means that the sea ice is melting and the coverage is shrinking which can have a global impact. Monitoring these sea ice changes are important to get an understanding of this impact. There are multiple options for how to carry out this monitoring, but the Arctic is a cold, dark and remote region, so a good option is using satellite images. A special advantage of using SAR (Synthetic Aperture Radar) satellites is that they are insensitive to weather conditions and darkness, which can be a challenge in the Arctic regions, especially during the winter months. Today operational ice services produces sea ice charts using such SAR images but they primarily reply on dual-polarimetric data. These images covers large areas due to the large swath for these modes and are idea for a Pan-Arctic coverage. However, they are impacted by e.g., effects of varying incidence angle throughout the images. This thesis will study the possibilities that lies within radar polarimetry, a field that utilized fully polarimetric data to extract even more information regarding the target on the surface, in this case sea ice. Moreover, these images enabled extraction of scattering type information and through this added information improve the sea ice type classification maps. The images studied here are two sets of fully polarimetric RADARSAT-2 data that covers the same sea ice but with varying incidence angles. Four different polarimetric features; the co-polarization ratio, the M-parameter, degree of polarization as well as the polarimetric coherence, were investigated for their separability measures between different sea ice types as well as how affected they are by incidence angle variations. Out of these four the co-polarization ratio was found to be useful to separate the thinner sea ice types as well as open water from the surrounding thicker sea ice. Both the M-parameter and the degree of polarization showed promise in separating deformed sea ice from the other ice types at higher incidence angles. Though limited consistent separability between different ice types were observed for the polarimetric coherence. The polarimetric coherence was found to be independent on the incidence angle, though the other three parameters showed an incidence angle dependency.
format Master Thesis
author Kinderås, Ida G.
author_facet Kinderås, Ida G.
author_sort Kinderås, Ida G.
title A Study of Polarimetric Parameters Used for Sea Ice Classification
title_short A Study of Polarimetric Parameters Used for Sea Ice Classification
title_full A Study of Polarimetric Parameters Used for Sea Ice Classification
title_fullStr A Study of Polarimetric Parameters Used for Sea Ice Classification
title_full_unstemmed A Study of Polarimetric Parameters Used for Sea Ice Classification
title_sort study of polarimetric parameters used for sea ice classification
publisher UiT The Arctic University of Norway
publishDate 2021
url https://hdl.handle.net/10037/25092
geographic Arctic
geographic_facet Arctic
genre Arctic
Sea ice
genre_facet Arctic
Sea ice
op_relation https://hdl.handle.net/10037/25092
op_rights openAccess
Copyright 2021 The Author(s)
_version_ 1766328434494013440