Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission
Several methods have been developed to provide polar maps of sea ice thickness (SIT) from L-band brightness temperature (TB) and altimetry data. Current process-based inversion methods to yield SIT fail to address the complex surface characteristics because sea ice is subject to strong seasonal dyna...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , |
Other Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute (MDPI)
2021
|
Subjects: | |
Online Access: | http://hdl.handle.net/2117/350116 https://doi.org/10.3390/rs13071366 |
id |
ftupcatalunyair:oai:upcommons.upc.edu:2117/350116 |
---|---|
record_format |
openpolar |
spelling |
ftupcatalunyair:oai:upcommons.upc.edu:2117/350116 2024-09-15T17:44:31+00:00 Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission Herbert, Christoph Josef Muñoz Martin, Joan Francesc Llaveria Godoy, David Pablos Hernández, Miriam Camps Carmona, Adriano José Universitat Politècnica de Catalunya. Doctorat en Teoria del Senyal i Comunicacions Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions Universitat Politècnica de Catalunya. RSLAB - Grup de Recerca en Teledetecció 2021-04-02 20 p. application/pdf http://hdl.handle.net/2117/350116 https://doi.org/10.3390/rs13071366 eng eng Multidisciplinary Digital Publishing Institute (MDPI) https://www.mdpi.com/2072-4292/13/7/1366 info:eu-repo/grantAgreement/FEDER/MDM-2016-0600 Herbert, C. [et al.]. Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission. "Remote sensing", 2 Abril 2021, vol. 13, núm. 7, p. 1-20. 2072-4292 http://hdl.handle.net/2117/350116 doi:10.3390/rs13071366 Attribution 3.0 Spain http://creativecommons.org/licenses/by/3.0/es/ Open Access Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Neural networks (Computer science) Microwave Predictive regression neural networks Sea Ice thickness Microwave radiometry Cubesats Xarxes neuronals (Informàtica) Microones Article 2021 ftupcatalunyair https://doi.org/10.3390/rs13071366 2024-07-25T11:13:07Z Several methods have been developed to provide polar maps of sea ice thickness (SIT) from L-band brightness temperature (TB) and altimetry data. Current process-based inversion methods to yield SIT fail to address the complex surface characteristics because sea ice is subject to strong seasonal dynamics and ice-physical properties are often non-linearly related. Neural networks can be trained to find hidden links among large datasets and often perform better on convoluted problems for which traditional approaches miss out important relationships between the observations. The FSSCat mission launched on 3 September 2020, carries the Flexible Microwave Payload-2 (FMPL-2), which contains the first Reflected Global Navigation Satellite System (GNSS-R) and L-band radiometer on board a CubeSat—designed to provide TB data on global coverage for soil moisture retrieval, and sea ice applications. This work investigates a predictive regression neural network approach with the goal to infer SIT using FMPL-2 TB and ancillary data (sea ice concentration, surface temperature, and sea ice freeboard). Two models—covering thin ice up to 0.6 m and full-range thickness—were separately trained on Arctic data in a two-month period from mid-October to the beginning of December 2020, while using ground truth data derived from the Soil Moisture and Ocean Salinity (SMOS) and Cryosat-2 missions. The thin ice and the full-range models resulted in a mean absolute error of 6.5 cm and 23 cm, respectively. Both of the models allowed for one to produce weekly composites of Arctic maps, and monthly composites of Antarctic SIT were predicted based on the Arctic full-range model. This work presents the first results of the FSSCat mission over the polar regions. It reveals the benefits of neural networks for sea ice retrievals and demonstrates that moderate-cost CubeSat missions can provide valuable data for applications in Earth observation. This work was supported by 2017 ESA S3 challenge and Copernicus Masters overall winner award (“FSSCat” ... Article in Journal/Newspaper Antarc* Antarctic Sea ice Universitat Politècnica de Catalunya, BarcelonaTech: UPCommons - Global access to UPC knowledge Remote Sensing 13 7 1366 |
institution |
Open Polar |
collection |
Universitat Politècnica de Catalunya, BarcelonaTech: UPCommons - Global access to UPC knowledge |
op_collection_id |
ftupcatalunyair |
language |
English |
topic |
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Neural networks (Computer science) Microwave Predictive regression neural networks Sea Ice thickness Microwave radiometry Cubesats Xarxes neuronals (Informàtica) Microones |
spellingShingle |
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Neural networks (Computer science) Microwave Predictive regression neural networks Sea Ice thickness Microwave radiometry Cubesats Xarxes neuronals (Informàtica) Microones Herbert, Christoph Josef Muñoz Martin, Joan Francesc Llaveria Godoy, David Pablos Hernández, Miriam Camps Carmona, Adriano José Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
topic_facet |
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació Neural networks (Computer science) Microwave Predictive regression neural networks Sea Ice thickness Microwave radiometry Cubesats Xarxes neuronals (Informàtica) Microones |
description |
Several methods have been developed to provide polar maps of sea ice thickness (SIT) from L-band brightness temperature (TB) and altimetry data. Current process-based inversion methods to yield SIT fail to address the complex surface characteristics because sea ice is subject to strong seasonal dynamics and ice-physical properties are often non-linearly related. Neural networks can be trained to find hidden links among large datasets and often perform better on convoluted problems for which traditional approaches miss out important relationships between the observations. The FSSCat mission launched on 3 September 2020, carries the Flexible Microwave Payload-2 (FMPL-2), which contains the first Reflected Global Navigation Satellite System (GNSS-R) and L-band radiometer on board a CubeSat—designed to provide TB data on global coverage for soil moisture retrieval, and sea ice applications. This work investigates a predictive regression neural network approach with the goal to infer SIT using FMPL-2 TB and ancillary data (sea ice concentration, surface temperature, and sea ice freeboard). Two models—covering thin ice up to 0.6 m and full-range thickness—were separately trained on Arctic data in a two-month period from mid-October to the beginning of December 2020, while using ground truth data derived from the Soil Moisture and Ocean Salinity (SMOS) and Cryosat-2 missions. The thin ice and the full-range models resulted in a mean absolute error of 6.5 cm and 23 cm, respectively. Both of the models allowed for one to produce weekly composites of Arctic maps, and monthly composites of Antarctic SIT were predicted based on the Arctic full-range model. This work presents the first results of the FSSCat mission over the polar regions. It reveals the benefits of neural networks for sea ice retrievals and demonstrates that moderate-cost CubeSat missions can provide valuable data for applications in Earth observation. This work was supported by 2017 ESA S3 challenge and Copernicus Masters overall winner award (“FSSCat” ... |
author2 |
Universitat Politècnica de Catalunya. Doctorat en Teoria del Senyal i Comunicacions Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions Universitat Politècnica de Catalunya. RSLAB - Grup de Recerca en Teledetecció |
format |
Article in Journal/Newspaper |
author |
Herbert, Christoph Josef Muñoz Martin, Joan Francesc Llaveria Godoy, David Pablos Hernández, Miriam Camps Carmona, Adriano José |
author_facet |
Herbert, Christoph Josef Muñoz Martin, Joan Francesc Llaveria Godoy, David Pablos Hernández, Miriam Camps Carmona, Adriano José |
author_sort |
Herbert, Christoph Josef |
title |
Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
title_short |
Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
title_full |
Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
title_fullStr |
Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
title_full_unstemmed |
Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission |
title_sort |
sea ice thickness estimation based on regression neural networks using l-band microwave radiometry data from the fsscat mission |
publisher |
Multidisciplinary Digital Publishing Institute (MDPI) |
publishDate |
2021 |
url |
http://hdl.handle.net/2117/350116 https://doi.org/10.3390/rs13071366 |
genre |
Antarc* Antarctic Sea ice |
genre_facet |
Antarc* Antarctic Sea ice |
op_relation |
https://www.mdpi.com/2072-4292/13/7/1366 info:eu-repo/grantAgreement/FEDER/MDM-2016-0600 Herbert, C. [et al.]. Sea ice thickness estimation based on regression neural networks using L-band microwave radiometry data from the FSSCat mission. "Remote sensing", 2 Abril 2021, vol. 13, núm. 7, p. 1-20. 2072-4292 http://hdl.handle.net/2117/350116 doi:10.3390/rs13071366 |
op_rights |
Attribution 3.0 Spain http://creativecommons.org/licenses/by/3.0/es/ Open Access |
op_doi |
https://doi.org/10.3390/rs13071366 |
container_title |
Remote Sensing |
container_volume |
13 |
container_issue |
7 |
container_start_page |
1366 |
_version_ |
1810492152150491136 |