CryoSat-2 waveform classification for melt event monitoring
Measuring the mass balance of ice sheets is important with respect to understanding among others sea level rise, glacier dynamics, global ocean circulation and marine ecosystems. One important parameter of the mass balance is surface melt, which can be estimated from different satellite data sources...
Published in: | Proceedings of the Northern Lights Deep Learning Workshop |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
2022
|
Subjects: | |
Online Access: | https://eprints.lancs.ac.uk/id/eprint/168915/ https://doi.org/10.7557/18.6284 |
id |
ftulancaster:oai:eprints.lancs.ac.uk:168915 |
---|---|
record_format |
openpolar |
spelling |
ftulancaster:oai:eprints.lancs.ac.uk:168915 2023-08-27T04:09:36+02:00 CryoSat-2 waveform classification for melt event monitoring Vermeer, Martijn Völgyes, David McMillan, Malcolm Fantin, Daniele 2022-03-28 https://eprints.lancs.ac.uk/id/eprint/168915/ https://doi.org/10.7557/18.6284 unknown Vermeer, Martijn and Völgyes, David and McMillan, Malcolm and Fantin, Daniele (2022) CryoSat-2 waveform classification for melt event monitoring. Proceedings of the Northern Lights Deep Learning Workshop, 3. ISSN 2703-6928 Journal Article PeerReviewed 2022 ftulancaster https://doi.org/10.7557/18.6284 2023-08-03T22:41:19Z Measuring the mass balance of ice sheets is important with respect to understanding among others sea level rise, glacier dynamics, global ocean circulation and marine ecosystems. One important parameter of the mass balance is surface melt, which can be estimated from different satellite data sources. In this study we investigate the potential of utilizing machine learning techniques for CryoSat-2 (CS2) radar altimeter waveform classification in order to derive melt information. Training data is derived by spatio-temporally matching of CS2 measurements with MODIS land surface temperature measurements. We propose a time convolution network with a fully connected classifier tail for CS2 waveform classifcation. In addition a non-deep learning model is implemented, providing a baseline. One of the main challenges is the high class imbalance, as surface temperatures on the interior of Greenland rarely reach the freezing point. The model performance is measured by several metrics: F1 score, average recall and Matthews correlation coefficient. The results of this proof of concept study indicate feasibility. Article in Journal/Newspaper glacier Greenland Lancaster University: Lancaster Eprints Greenland Proceedings of the Northern Lights Deep Learning Workshop 3 |
institution |
Open Polar |
collection |
Lancaster University: Lancaster Eprints |
op_collection_id |
ftulancaster |
language |
unknown |
description |
Measuring the mass balance of ice sheets is important with respect to understanding among others sea level rise, glacier dynamics, global ocean circulation and marine ecosystems. One important parameter of the mass balance is surface melt, which can be estimated from different satellite data sources. In this study we investigate the potential of utilizing machine learning techniques for CryoSat-2 (CS2) radar altimeter waveform classification in order to derive melt information. Training data is derived by spatio-temporally matching of CS2 measurements with MODIS land surface temperature measurements. We propose a time convolution network with a fully connected classifier tail for CS2 waveform classifcation. In addition a non-deep learning model is implemented, providing a baseline. One of the main challenges is the high class imbalance, as surface temperatures on the interior of Greenland rarely reach the freezing point. The model performance is measured by several metrics: F1 score, average recall and Matthews correlation coefficient. The results of this proof of concept study indicate feasibility. |
format |
Article in Journal/Newspaper |
author |
Vermeer, Martijn Völgyes, David McMillan, Malcolm Fantin, Daniele |
spellingShingle |
Vermeer, Martijn Völgyes, David McMillan, Malcolm Fantin, Daniele CryoSat-2 waveform classification for melt event monitoring |
author_facet |
Vermeer, Martijn Völgyes, David McMillan, Malcolm Fantin, Daniele |
author_sort |
Vermeer, Martijn |
title |
CryoSat-2 waveform classification for melt event monitoring |
title_short |
CryoSat-2 waveform classification for melt event monitoring |
title_full |
CryoSat-2 waveform classification for melt event monitoring |
title_fullStr |
CryoSat-2 waveform classification for melt event monitoring |
title_full_unstemmed |
CryoSat-2 waveform classification for melt event monitoring |
title_sort |
cryosat-2 waveform classification for melt event monitoring |
publishDate |
2022 |
url |
https://eprints.lancs.ac.uk/id/eprint/168915/ https://doi.org/10.7557/18.6284 |
geographic |
Greenland |
geographic_facet |
Greenland |
genre |
glacier Greenland |
genre_facet |
glacier Greenland |
op_relation |
Vermeer, Martijn and Völgyes, David and McMillan, Malcolm and Fantin, Daniele (2022) CryoSat-2 waveform classification for melt event monitoring. Proceedings of the Northern Lights Deep Learning Workshop, 3. ISSN 2703-6928 |
op_doi |
https://doi.org/10.7557/18.6284 |
container_title |
Proceedings of the Northern Lights Deep Learning Workshop |
container_volume |
3 |
_version_ |
1775351100302426112 |