Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet
Climate change is extensively affecting ice sheets resulting in accelerating mass loss in recent decades. Assessment of this reduction and its causes is required to project future ice mass loss. Annual snow accumulation is an important component of the surface mass balance of ice sheets. While in si...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2021
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs13142707 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/13/14/2707/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/13/14/2707/ 2023-08-20T04:06:53+02:00 Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet Debvrat Varshney Maryam Rahnemoonfar Masoud Yari John Paden Oluwanisola Ibikunle Jilu Li agris 2021-07-09 application/pdf https://doi.org/10.3390/rs13142707 EN eng Multidisciplinary Digital Publishing Institute Biogeosciences Remote Sensing https://dx.doi.org/10.3390/rs13142707 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 13; Issue 14; Pages: 2707 ice layer thickness semantic segmentation fully convolutional networks Snow Radar Text 2021 ftmdpi https://doi.org/10.3390/rs13142707 2023-08-01T02:09:13Z Climate change is extensively affecting ice sheets resulting in accelerating mass loss in recent decades. Assessment of this reduction and its causes is required to project future ice mass loss. Annual snow accumulation is an important component of the surface mass balance of ice sheets. While in situ snow accumulation measurements are temporally and spatially limited due to their high cost, airborne radar sounders can achieve ice sheet wide coverage by capturing and tracking annual snow layers in the radar images or echograms. In this paper, we use deep learning to uniquely identify the position of each annual snow layer in the Snow Radar echograms taken across different regions over the Greenland ice sheet. We train with more than 15,000 images generated from radar echograms and estimate the thickness of each snow layer within a mean absolute error of 0.54 to 7.28 pixels, depending on dataset. A highly precise snow layer thickness can help improve weather models and, thus, support glaciological studies. Such a well-trained deep learning model can be used with ever-growing datasets to aid in the accurate assessment of snow accumulation on the dynamically changing ice sheets. Text Greenland Ice Sheet MDPI Open Access Publishing Greenland Remote Sensing 13 14 2707 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
ice layer thickness semantic segmentation fully convolutional networks Snow Radar |
spellingShingle |
ice layer thickness semantic segmentation fully convolutional networks Snow Radar Debvrat Varshney Maryam Rahnemoonfar Masoud Yari John Paden Oluwanisola Ibikunle Jilu Li Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
topic_facet |
ice layer thickness semantic segmentation fully convolutional networks Snow Radar |
description |
Climate change is extensively affecting ice sheets resulting in accelerating mass loss in recent decades. Assessment of this reduction and its causes is required to project future ice mass loss. Annual snow accumulation is an important component of the surface mass balance of ice sheets. While in situ snow accumulation measurements are temporally and spatially limited due to their high cost, airborne radar sounders can achieve ice sheet wide coverage by capturing and tracking annual snow layers in the radar images or echograms. In this paper, we use deep learning to uniquely identify the position of each annual snow layer in the Snow Radar echograms taken across different regions over the Greenland ice sheet. We train with more than 15,000 images generated from radar echograms and estimate the thickness of each snow layer within a mean absolute error of 0.54 to 7.28 pixels, depending on dataset. A highly precise snow layer thickness can help improve weather models and, thus, support glaciological studies. Such a well-trained deep learning model can be used with ever-growing datasets to aid in the accurate assessment of snow accumulation on the dynamically changing ice sheets. |
format |
Text |
author |
Debvrat Varshney Maryam Rahnemoonfar Masoud Yari John Paden Oluwanisola Ibikunle Jilu Li |
author_facet |
Debvrat Varshney Maryam Rahnemoonfar Masoud Yari John Paden Oluwanisola Ibikunle Jilu Li |
author_sort |
Debvrat Varshney |
title |
Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
title_short |
Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
title_full |
Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
title_fullStr |
Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
title_full_unstemmed |
Deep Learning on Airborne Radar Echograms for Tracing Snow Accumulation Layers of the Greenland Ice Sheet |
title_sort |
deep learning on airborne radar echograms for tracing snow accumulation layers of the greenland ice sheet |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2021 |
url |
https://doi.org/10.3390/rs13142707 |
op_coverage |
agris |
geographic |
Greenland |
geographic_facet |
Greenland |
genre |
Greenland Ice Sheet |
genre_facet |
Greenland Ice Sheet |
op_source |
Remote Sensing; Volume 13; Issue 14; Pages: 2707 |
op_relation |
Biogeosciences Remote Sensing https://dx.doi.org/10.3390/rs13142707 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs13142707 |
container_title |
Remote Sensing |
container_volume |
13 |
container_issue |
14 |
container_start_page |
2707 |
_version_ |
1774718239148867584 |