Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)

Increase in the global mean temperature is extensively affecting ice sheets and reducing them. Assessment of this reduction and its causes is required to project its global climatic impact. A usual way of analyzing the reduction is by calculating the change in annual snow accumulation over these ice...

Full description

Bibliographic Details
Main Authors: Debvrat Varshney (10942074), Maryam Rahnemoonfar (9739398), Masoud Yari (10942165), John Paden (10942181), Ibikunle Oluwanisola (10954760), Jilu Li (10954769)
Format: Still Image
Language:unknown
Published: 2021
Subjects:
Online Access:https://doi.org/10.6084/m9.figshare.14765541.v2
id ftsmithonian:oai:figshare.com:article/14765541
record_format openpolar
spelling ftsmithonian:oai:figshare.com:article/14765541 2023-05-15T16:26:51+02:00 Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017) Debvrat Varshney (10942074) Maryam Rahnemoonfar (9739398) Masoud Yari (10942165) John Paden (10942181) Ibikunle Oluwanisola (10954760) Jilu Li (10954769) 2021-06-10T17:49:27Z https://doi.org/10.6084/m9.figshare.14765541.v2 unknown https://figshare.com/articles/poster/varshney_debvrat_png/14765541 doi:10.6084/m9.figshare.14765541.v2 CC BY 4.0 CC-BY Climate Change Processes Glaciology Ice Layer Thickness Semantic Segmentation Fully Convolutional Networks Snow Radar Greenland Image Poster 2021 ftsmithonian https://doi.org/10.6084/m9.figshare.14765541.v2 2021-06-13T14:38:50Z Increase in the global mean temperature is extensively affecting ice sheets and reducing them. Assessment of this reduction and its causes is required to project its global climatic impact. A usual way of analyzing the reduction is by calculating the change in annual snow accumulation over these ice sheets. This can be done with the help of images from the Snow Radar sensor, an airborne sensor which captures the snow profile of different years. In this paper, we use deep learning to uniquely identify the position of each annual snow layer in the Snow Radar images taken across different regions over the Greenland ice sheet. We train with more than 15,000 images and estimate the thickness of each snow layer within a mean absolute error of 0.54 to 7.28 pixels, depending on year. A highly precise snow layer thickness can help study the ablation and accumulation of snow through time 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. Still Image Greenland Ice Sheet Unknown Greenland
institution Open Polar
collection Unknown
op_collection_id ftsmithonian
language unknown
topic Climate Change Processes
Glaciology
Ice Layer Thickness
Semantic Segmentation
Fully Convolutional Networks
Snow Radar
Greenland
spellingShingle Climate Change Processes
Glaciology
Ice Layer Thickness
Semantic Segmentation
Fully Convolutional Networks
Snow Radar
Greenland
Debvrat Varshney (10942074)
Maryam Rahnemoonfar (9739398)
Masoud Yari (10942165)
John Paden (10942181)
Ibikunle Oluwanisola (10954760)
Jilu Li (10954769)
Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
topic_facet Climate Change Processes
Glaciology
Ice Layer Thickness
Semantic Segmentation
Fully Convolutional Networks
Snow Radar
Greenland
description Increase in the global mean temperature is extensively affecting ice sheets and reducing them. Assessment of this reduction and its causes is required to project its global climatic impact. A usual way of analyzing the reduction is by calculating the change in annual snow accumulation over these ice sheets. This can be done with the help of images from the Snow Radar sensor, an airborne sensor which captures the snow profile of different years. In this paper, we use deep learning to uniquely identify the position of each annual snow layer in the Snow Radar images taken across different regions over the Greenland ice sheet. We train with more than 15,000 images and estimate the thickness of each snow layer within a mean absolute error of 0.54 to 7.28 pixels, depending on year. A highly precise snow layer thickness can help study the ablation and accumulation of snow through time 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 Still Image
author Debvrat Varshney (10942074)
Maryam Rahnemoonfar (9739398)
Masoud Yari (10942165)
John Paden (10942181)
Ibikunle Oluwanisola (10954760)
Jilu Li (10954769)
author_facet Debvrat Varshney (10942074)
Maryam Rahnemoonfar (9739398)
Masoud Yari (10942165)
John Paden (10942181)
Ibikunle Oluwanisola (10954760)
Jilu Li (10954769)
author_sort Debvrat Varshney (10942074)
title Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
title_short Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
title_full Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
title_fullStr Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
title_full_unstemmed Deep radiostratigraphy of Greenland Ice sheet through Deep Learning on airborne Snow Radar images (2009-2017)
title_sort deep radiostratigraphy of greenland ice sheet through deep learning on airborne snow radar images (2009-2017)
publishDate 2021
url https://doi.org/10.6084/m9.figshare.14765541.v2
geographic Greenland
geographic_facet Greenland
genre Greenland
Ice Sheet
genre_facet Greenland
Ice Sheet
op_relation https://figshare.com/articles/poster/varshney_debvrat_png/14765541
doi:10.6084/m9.figshare.14765541.v2
op_rights CC BY 4.0
op_rightsnorm CC-BY
op_doi https://doi.org/10.6084/m9.figshare.14765541.v2
_version_ 1766015850791305216