Home > Research > Publications & Outputs > CryoSat-2 waveform classification for melt even...

Research

Links

Text available via DOI:

Keywords

View graph of relations

CryoSat-2 waveform classification for melt event monitoring

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

CryoSat-2 waveform classification for melt event monitoring. / Vermeer, Martijn; Völgyes, David; McMillan, Malcolm et al.
In: Proceedings of the Northern Lights Deep Learning Workshop, Vol. 3, 28.03.2022.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Vermeer, M, Völgyes, D, McMillan, M & Fantin, D 2022, 'CryoSat-2 waveform classification for melt event monitoring', Proceedings of the Northern Lights Deep Learning Workshop, vol. 3. https://doi.org/10.7557/18.6284

APA

Vermeer, M., Völgyes, D., McMillan, M., & Fantin, D. (2022). CryoSat-2 waveform classification for melt event monitoring. Proceedings of the Northern Lights Deep Learning Workshop, 3. https://doi.org/10.7557/18.6284

Vancouver

Vermeer M, Völgyes D, McMillan M, Fantin D. CryoSat-2 waveform classification for melt event monitoring. Proceedings of the Northern Lights Deep Learning Workshop. 2022 Mar 28;3. doi: 10.7557/18.6284

Author

Vermeer, Martijn ; Völgyes, David ; McMillan, Malcolm et al. / CryoSat-2 waveform classification for melt event monitoring. In: Proceedings of the Northern Lights Deep Learning Workshop. 2022 ; Vol. 3.

Bibtex

@article{0bc5ffc6e9a44da5b64bcfb366377341,
title = "CryoSat-2 waveform classification for melt event monitoring",
abstract = "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.",
keywords = "deep learning, surface mass balance, melt dynamics, Greenland, ice sheet, Cryosat-2, MODIS",
author = "Martijn Vermeer and David V{\"o}lgyes and Malcolm McMillan and Daniele Fantin",
year = "2022",
month = mar,
day = "28",
doi = "10.7557/18.6284",
language = "English",
volume = "3",
journal = "Proceedings of the Northern Lights Deep Learning Workshop",
issn = "2703-6928",
publisher = "UiT The Arctic University of Norway",

}

RIS

TY - JOUR

T1 - CryoSat-2 waveform classification for melt event monitoring

AU - Vermeer, Martijn

AU - Völgyes, David

AU - McMillan, Malcolm

AU - Fantin, Daniele

PY - 2022/3/28

Y1 - 2022/3/28

N2 - 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.

AB - 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.

KW - deep learning

KW - surface mass balance

KW - melt dynamics

KW - Greenland

KW - ice sheet

KW - Cryosat-2

KW - MODIS

U2 - 10.7557/18.6284

DO - 10.7557/18.6284

M3 - Journal article

VL - 3

JO - Proceedings of the Northern Lights Deep Learning Workshop

JF - Proceedings of the Northern Lights Deep Learning Workshop

SN - 2703-6928

ER -