Combined Use of Space Borne Optical and SAR Data to Improve Knowledge about Sea Ice for Shipping
Abstract
:1. Introduction
2. Materials and Methods
2.1. Algorithm for Sea Ice Discrimination Using SLSTR Optical Data
2.2. Algorithm for Sea Ice Discrimination Using Sentinel-1 SAR Data
2.3. Approaches to Combine Sentinel-3 SLSTR and Sentinel-1 SAR Ice Classifications
3. Results
3.1. Verification of the Sentinel-3 SLSTR Ice Classification Results
3.2. First Approach to Automatically Combine Sentinel-3 SLSTR and Sentinel-1 SAR Ice Classifications
3.3. Second Approach to Combine Sentinel-3 SLSTR and Sentinel-1 SAR Ice Classifications
4. Discussion
4.1. Optical Ice Classification Algorithm Conversion
4.2. Optical—SAR Ice Classification Algorithm Combination
- open water, or nilas up to 10 cm thick, or flooded ice, displayed in blue;
- young ice: grey ice, or grey-white ice up to 30 cm thick, displayed in yellow;
- ice between 30 cm and 200 cm thick, displayed in magenta.
4.3. Further Plans to Improve the Combination of Sentinel-3 SLSTR and Sentinel-1 SAR Ice Classifications
5. Conclusions
- adapting and improving a sea ice classification based on NOAA AVHRR to work with Sentinel-3 SLSTR data;
- verifying the SLSTR based algorithm by expert’s visual inspection, comparison to ice maps available, as well as higher resolution satellite images adapting a sea ice classification algorithm based on RADARSAT2 or TerraSAR-X SAR data to work with Sentinel-1 SAR data;
- demonstrating that additional information is gained even by applying simple approaches to combine the results of two algorithms mentioned above;
- proposing a neural network for a more sophisticated combination to be investigated in future.
Author Contributions
Funding
Institutional Review Board Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Dey, B. Applications of Satellite Thermal Infrared Images for Monitoring North Water during the Periods of Polar Darkness. J. Glaciol. 1980, 25, 425–438. [Google Scholar] [CrossRef] [Green Version]
- Aguirre, M.; Berruti, B.; Bezy, J.; Drinkwater, M.; Heliere, F.; Klein, U.; Mavrocordatos, C.; Silvestrin, P.; Greco, B.; Benveniste, J. Sentinel-3—The Ocean and Medium-Resolution Land Mission for GMES Operational Services. ESA Bull. 2007, 131, 29. [Google Scholar]
- Donlon, C.; Berruti, B.; Buongiorno, A.; Ferreira, M.-H.; Féménias, P.; Frerick, J.; Goryl, P.; Klein, U.; Laur, H.; Mavrocordatos, C.; et al. The Global Monitoring for Environment and Security (GMES) Sentinel-3 mission. Remote Sens. Environ. 2012, 120, 37–57. [Google Scholar] [CrossRef]
- Riggs, A.G.; Hall, D.K.; Ackerman, A.S. Sea Ice Extent and Classification Mapping with the Moderate Resolution Imaging Spectroradiometer Airborne Simulator. Remote Sens. Environ. 1999, 68, 152–163. [Google Scholar] [CrossRef]
- König, C. Eisfernerkundung mit NOAA-AVHRR und SAR. Dtsch. Hydrogr. Z. 1995, 4. ISSN 0946-2015. Available online: https://openlibrary.org/books/OL76781M/Eisfernerkundung_mit_NOAA-advanced_very_high_resolution_radiometer_%28AVHRR%29_und_synthetic_aperture_ra (accessed on 23 November 2021).
- Kidwell, K. (Ed.) NOAA Polar Orbiter Data User’s Guide (TIROS-N, NOAA-6, NOAA-7, NOAA-8, NOAA-9, NOAA-10, NOAA-11, NOAA-12, NOAA-13 and NOAA-14). NOAA NESDIS, November 1998. Available online: https://web.archive.org/web/20161209021928/https://www.ncdc.noaa.gov/oa/pod-guide/ncdc/docs/podug/index.htm (accessed on 23 November 2021).
- Robel, J.; Graumann, A. (Eds.) NOAA KLM Users’s Guide with NOAA-N, N Prime, and MetOp Supplements. NOAA NESDIS, April 2014. p. 2530. Available online: https://www.star.nesdis.noaa.gov/mirs/documents/0.0_NOAA_KLM_Users_Guide.pdf (accessed on 23 November 2021).
- Ressel, R.; Frost, A.; Lehner, S. A Neural Network-Based Classification for Sea Ice Types on X-Band SAR Images. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2015, 8, 3672–3680. [Google Scholar] [CrossRef] [Green Version]
- Singha, S. Towards Pan-Arctic Sea Ice Type Retrieval Using Sentinel-1 TOPSAR Modes. In Proceedings of the EUSAR 2021—13th European Conference on Synthetic Aperture Radar, Online Event, 29 March–1 April 2021; Available online: https://www.vde-verlag.de/proceedings-de/455457180.html (accessed on 23 November 2021).
- Park, J.-W.; Korosov, A.A.; Babiker, M.; Sandven, S.; Won, J.-S. Efficient Thermal Noise Removal for Sentinel-1 TOPSAR Cross-Polarization Channel. IEEE Trans. Geosci. Remote Sens. 2017, 56, 1555–1565. [Google Scholar] [CrossRef]
- Jolliffe, I. Principal Component Analysis, 2nd ed.; Springer: New York, NY, USA, 2002; Available online: http://cda.psych.uiuc.edu/statistical_learning_course/Jolliffe%20I.%20Principal%20Component%20Analysis%20(2ed.,%20Springer,%202002)(518s)_MVsa_.pdf (accessed on 23 November 2021).
- König, C.; König, T.; Frost, A.; Jacobsen, S. Verbesserung der Meereis-Lageinformationen für die Schifffahrt in Polaren Gewässern Durch Kombinierte Meereis-Klassifikation mit optischen Daten der Sentinel-3 und SAR-Daten der Sentinel-1 Satellitenserie." Final Report of Project EisKlass31 (in German), Project Timeframe: August 2018—July 2019, Submitted to: Technische Informationsbibliothek Hannover. 2019. Available online: https://www.tib.eu/de/suchen/id/TIBKAT:1767649789/Verbesserung-der-Meereis-Lageinformationen-f%C3%BCr?cHash=2af0a7aa5ec710dffa6a7553d8a8a58f (accessed on 23 November 2021).
AVHRR | SLSTR | ||
---|---|---|---|
Channel 1 | 0.58–0.68 µm | Channel S1 | 0.545–0.565 µm |
Channel S2 | 0.649–0.669 µm | ||
Channel 2 | 0.725–1.00 µm | Channel S3 | 0.858–0.878 µm |
Channel S4 | 1.364–1.384 µm | ||
Channel 3A | 1.6 µm | Channel S5 | 1.583–1.643 µm |
Channel 3B | 3.5–3.9 µm | Channel S7 | 3.4–4.0 µm |
Channel 4 | 10.3–11.3 µm | Channel S8 | 10.5–11.4 µm |
Channel 5 | 11.5–12.5 µm | Channel S9 | 11.0–13.0 µm |
Colour Name and Image of Central Colour in Class | Mean Reflectivity (%) Channel S2 | Mean Brightness Temperature (°C) Channel S9 | Interpretation |
---|---|---|---|
green | 4.7 | 1.3 | Open water |
green-orange | 3.0 | −3.2 | nearly freezing water, or mixed ice/water pixel or very thin ice (e.g., frazil ice) |
bright and dark orange | 10.4 | −7.5 | ice without snow cover; especially dark and light nilas (<10 cm) |
medium yellow | 28.9 | −10.9 | young ice without snow cover; especially grey ice (10–15 cm) |
light yellow-white | 34.8 | −11.6 | young ice without snow cover; especially grey-white ice (15–30 cm) |
dark red | 69.7 | −20.3 | ice of uncertain thickness with dry snow cover |
medium red | 69.8 | −21.6 | ice of uncertain thickness with aged dry snow cover |
light red | 75.5 | −24.2 | ice of uncertain thickness with aged snow cover |
dark pink | 80.4 | −15.5 | ice of uncertain thickness (most likely between 20 and 50 cm) covered by aged snow (increased grain size of snow) |
medium pink | 82.2 | −21.0 | ice of uncertain thickness (most likely between 20 and 50 cm) covered by further aged snow compared to dark pink |
light pink | 83.6 | −18.6 | ice of uncertain thickness (most likely between 20 and 50 cm) covered by snow |
dark violet | 84.4 | −12.9 | ice of uncertain thickness covered by thick, slightly aged snow with very small amount of moisture |
light violet | 82.1 | −11.9 | similar to dark violet, but advanced snow metamorphism and moisture |
light blue | 68.7 | −1.2 | thick ice covered by slightly wet snow (thickness not well defined) |
dark blue | 52.3 | 0.0 | Ice covered by increasingly wet snow; partial pixel coverage by meltponds possible |
light green | 26.5 | −2.3 | Ice with low snow cover and mainly meltponds, probably refrozen on the surface |
SAR Ice Class Colour | |||
---|---|---|---|
SLSTR Ice Class Colour and Ice Class | Colour Used for Combination of SAR Class 1 (Open Water or Nilas up to 10 cm) and SLSTR Class According to Table Row | Colour Used for Combination of SAR Class 2 (Young Ice, 10–30 cm) and SLSTR Class According to Table Row | Colour Used for Combination of SAR Class 3 (Smooth First Year Ice, 30–200 cm) and SLSTR class According to Table Row |
open water | open water | open water | inconsistent |
freezing ready water | freezing ready water | inconsistent | inconsistent |
nilas: 0–10 cm | nilas | nilas | inconsistent |
grey ice: 10–15 cm | inconsistent | grey ice up to 15 cm | inconsistent |
grey-white ice: 15–30 cm | inconsistent | grey-white ice | smooth grey-white ice near 30 cm |
dark red: Ice of unknown thickness covered by dry snow | nilas covered by dry snow | young ice 10–30 cm covered by dry snow | smooth first year ice, 30–200 cm covered by dry snow |
medium red: Ice of unknown thickness covered by aged snow | inconsistent | young ice, 10–30 cm covered by aged snow | smooth first year ice, 30–200 cm covered by aged snow |
light red; Ice of unknown thickness covered by less aged snow | inconsistent | young ice, 10–30 cm covered by less aged snow | smooth first year ice, 30–200 cm covered by less aged snow |
dark pink: Ice covered by aged snow at low temperature | inconsistent | young ice, 10–30 cm covered by aged cold snow | smooth first year ice, 30–200 cm covered by aged cold snow |
medium pink: Ice covered by aged snow at low temperature | inconsistent | young ice, 10–30 cm covered by aged cold snow | smooth first year ice, 30–200 cm covered by aged cold snow |
light pink: Ice covered by aged snow at low temperature | inconsistent | young ice, 10–30 cm covered by aged cold snow | smooth first year ice, 30–200 cm covered by aged cold snow |
dark violet: Ice of uncertain thickness covered by thick, slightly aged snow | inconsistent | young ice, 10–30 cm covered by slightly aged snow | smooth first year ice, 30–200 cm covered by slightly aged snow |
light violet: Similar to dark violet, but advanced snow metamorphism | inconsistent | young ice, 10–30 cm covered by aged snow | smooth first year ice, 30–200 cm covered by aged snow |
light blue: Thick ice covered by moist snow | nilas covered by moist snow | young ice, 10–30 cm covered by moist snow | smooth first year ice, 30–200 cm covered by moist snow |
dark blue: Ice covered by increasingly wet snow; partial cover by meltponds possible | inconsistent | young ice, 10–30 cm covered by increasingly wet snow | smooth first year ice, 30–200 cm covered by increasingly wet snow |
light green: Ice with low snow cover and mainly meltponds, probably refrozen on the surface | inconsistent | young ice, 10–30 cm, probably refrozen on the surface | smooth first year ice, 30–200 cm, probably refrozen on the surface |
diluting ice (brownish), ice/water mixture | ice/water mixture | ice/water mixture, ice up to 30cm | inconsistent |
diluting ice (greenish), ice/water mixture | ice/water mixture | ice/water mixture, ice up to 30cm | inconsistent |
light grey-white ice | inconsistent | grey-white ice up to 30 cm, very bright | smooth grey-white ice, at least 30 cm thick |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
König, C.; König, T.; Singha, S.; Frost, A.; Jacobsen, S. Combined Use of Space Borne Optical and SAR Data to Improve Knowledge about Sea Ice for Shipping. Remote Sens. 2021, 13, 4842. https://doi.org/10.3390/rs13234842
König C, König T, Singha S, Frost A, Jacobsen S. Combined Use of Space Borne Optical and SAR Data to Improve Knowledge about Sea Ice for Shipping. Remote Sensing. 2021; 13(23):4842. https://doi.org/10.3390/rs13234842
Chicago/Turabian StyleKönig, Christine, Thomas König, Suman Singha, Anja Frost, and Sven Jacobsen. 2021. "Combined Use of Space Borne Optical and SAR Data to Improve Knowledge about Sea Ice for Shipping" Remote Sensing 13, no. 23: 4842. https://doi.org/10.3390/rs13234842