Articles | Volume 18, issue 8
https://doi.org/10.5194/tc-18-3471-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Special issue:
https://doi.org/10.5194/tc-18-3471-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
The AutoICE Challenge
Andreas Stokholm
CORRESPONDING AUTHOR
DTU Space, Department of Space Research and Technology, Technical University of Denmark (DTU), Kgs. Lyngby, Denmark
ϕ-lab, European Space Research Institute (ESRIN), European Space Agency (ESA), Frascati, Italy
Jørgen Buus-Hinkler
Danish Meteorological Institute (DMI), Copenhagen, Denmark
Tore Wulf
Danish Meteorological Institute (DMI), Copenhagen, Denmark
Anton Korosov
Nansen Environmental and Remote Sensing Center (NERSC), Bergen, Norway
Roberto Saldo
DTU Space, Department of Space Research and Technology, Technical University of Denmark (DTU), Kgs. Lyngby, Denmark
Leif Toudal Pedersen
DTU Space, Department of Space Research and Technology, Technical University of Denmark (DTU), Kgs. Lyngby, Denmark
David Arthurs
Polar View, Herlev, Denmark
Ionut Dragan
SpaceTec Partners, Brussels, Belgium
Iacopo Modica
GMATICS, Rome, Italy
Juan Pedro
EarthPulse, Barcelona, Spain
Annekatrien Debien
SpaceTec Partners, Brussels, Belgium
Xinwei Chen
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Muhammed Patel
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Fernando Jose Pena Cantu
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Javier Noa Turnes
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Jinman Park
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Linlin Xu
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Katharine Andrea Scott
Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, Canada
David Anthony Clausi
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Yuan Fang
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Mingzhe Jiang
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Saeid Taleghanidoozdoozan
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Neil Curtis Brubacher
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Armina Soleymani
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Zacharie Gousseau
Department of System Design Engineering, University of Waterloo, Waterloo, Canada
Michał Smaczny
Warsaw University of Technology, Warsaw, Poland
Patryk Kowalski
Warsaw University of Technology, Warsaw, Poland
Jacek Komorowski
Warsaw University of Technology, Warsaw, Poland
David Rijlaarsdam
Ubotica Technologies, Dublin, Ireland
Jan Nicolaas van Rijn
Leiden Institute of Advanced Computer Science, Leiden University, Leiden, the Netherlands
Jens Jakobsen
Danish Meteorological Institute (DMI), Copenhagen, Denmark
Martin Samuel James Rogers
AI Lab, British Antarctic Survey, Cambridge, United Kingdom
Nick Hughes
Norwegian Ice Service, Norwegian Meteorological Institute, Oslo, Norway
Tom Zagon
Canadian Ice Service, Environment and Climate Change Canada, Ottawa, Canada
Rune Solberg
Norwegian Computing Center (NR), Oslo, Norway
Nicolas Longépé
ϕ-lab, European Space Research Institute (ESRIN), European Space Agency (ESA), Frascati, Italy
Matilde Brandt Kreiner
Danish Meteorological Institute (DMI), Copenhagen, Denmark
Related authors
Tim Enzlberger Jensen, Bjørnar Dale, Andreas Stokholm, René Forsberg, Alexandre Bresson, Nassim Zahzam, Alexis Bonnin, and Yannick Bidel
Earth Syst. Sci. Data, 17, 1667–1684, https://doi.org/10.5194/essd-17-1667-2025, https://doi.org/10.5194/essd-17-1667-2025, 2025
Short summary
Short summary
The availability of data from two airborne gravity campaigns using sensors based on both classical and quantum technology is presented. Data are made available by the European Space Agency as raw, intermediate, and final data products. Here “raw” refers to the sensor output, while “final” refers to the along-track gravity estimates. This makes the data relevant for users interested in applications ranging from data processing and quantum studies to geophysical studies using gravity observations.
Andreas Stokholm, Andrzej Kucik, Nicolas Longépé, and Sine Munk Hvidegaard
EGUsphere, https://doi.org/10.5194/egusphere-2023-976, https://doi.org/10.5194/egusphere-2023-976, 2023
Preprint archived
Short summary
Short summary
To navigate in the Arctic, vessels utilise maps of the local sea ice conditions, that are manually drawn daily. This is a time-consuming and labour-intensive task, which we are trying to automate using convolutional neural networks. In this article, we investigate combining the outputs of models trained with different objectives, leveraging the strengths and avoiding weaknesses to create precise and detailed sea ice concentration maps for navigation in the Arctic and climate or weather models.
Anton Korosov, Léo Edel, Heather Regan, Thomas Lavergne, Emily Jane Down, and Signe Aaboe
Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2025-477, https://doi.org/10.5194/essd-2025-477, 2025
Preprint under review for ESSD
Short summary
Short summary
We present a new long-term record of Arctic sea ice age spanning from 1991 to 2024. Using satellite data and a new tracking method, it maps fractions of sea ice from first- to sixth-year and includes uncertainty estimates. The dataset shows a decline in older ice and more first-year ice, it agrees well with buoy data, and supports Arctic monitoring, climate research, navigation, and model evaluation.
Tim Enzlberger Jensen, Bjørnar Dale, Andreas Stokholm, René Forsberg, Alexandre Bresson, Nassim Zahzam, Alexis Bonnin, and Yannick Bidel
Earth Syst. Sci. Data, 17, 1667–1684, https://doi.org/10.5194/essd-17-1667-2025, https://doi.org/10.5194/essd-17-1667-2025, 2025
Short summary
Short summary
The availability of data from two airborne gravity campaigns using sensors based on both classical and quantum technology is presented. Data are made available by the European Space Agency as raw, intermediate, and final data products. Here “raw” refers to the sensor output, while “final” refers to the along-track gravity estimates. This makes the data relevant for users interested in applications ranging from data processing and quantum studies to geophysical studies using gravity observations.
Mohammad Sina Jahangir, John Quilty, Chaopeng Shen, Andrea Scott, Scott Steinschneider, and Jan Adamowski
EGUsphere, https://doi.org/10.5194/egusphere-2025-846, https://doi.org/10.5194/egusphere-2025-846, 2025
Short summary
Short summary
This study presents a novel hybrid approach to streamflow prediction, significantly improving the efficiency and accuracy of fine-tuning deep learning models for hydrological prediction. Tested across numerous catchments in the U.S. and Europe, this method accelerates the fine-tuning process and improves prediction accuracy in locations beyond the training data. This innovative approach sets the stage for future hydrological models leveraging transfer learning.
Léo Edel, Jiping Xie, Anton Korosov, Julien Brajard, and Laurent Bertino
The Cryosphere, 19, 731–752, https://doi.org/10.5194/tc-19-731-2025, https://doi.org/10.5194/tc-19-731-2025, 2025
Short summary
Short summary
This study developed a new method to estimate Arctic sea ice thickness from 1992 to 2010 using a combination of machine learning and data assimilation. By training a machine learning model on data from 2011 to 2022, past errors in sea ice thickness can be corrected, leading to improved estimations. This approach provides insights into historical changes in sea ice thickness, showing a notable decline from 1992 to 2022, and offers a valuable resource for future studies.
Anton Korosov, Yue Ying, and Einar Ólason
Geosci. Model Dev., 18, 885–904, https://doi.org/10.5194/gmd-18-885-2025, https://doi.org/10.5194/gmd-18-885-2025, 2025
Short summary
Short summary
We have developed a new method to improve the accuracy of sea ice models, which predict how ice moves and deforms due to wind and ocean currents. Traditional models use parameters that are often poorly defined. The new approach uses machine learning to fine-tune these parameters by comparing simulated ice drift with satellite data. The method identifies optimal settings for the model by analysing patterns in ice deformation. This results in more accurate simulations of sea ice drift forecasting.
Einar Ólason, Guillaume Boutin, Timothy Williams, Anton Korosov, Heather Regan, Jonathan Rheinlænder, Pierre Rampal, Daniela Flocco, Abdoulaye Samaké, Richard Davy, Timothy Spain, and Sean Chua
EGUsphere, https://doi.org/10.5194/egusphere-2024-3521, https://doi.org/10.5194/egusphere-2024-3521, 2025
Short summary
Short summary
This paper introduces a new version of the neXtSIM sea-ice model. NeXtSIM is unique among sea-ice models in how it represents sea-ice dynamics, focusing on features such as cracks and ridges and how these impact interactions between the atmosphere and ocean where sea ice is present. The new version introduces some physical parameterisations and model options detailed and explained in the paper. Following the paper's publication, the neXtSIM code will be released publicly for the first time.
Tore Wulf, Jørgen Buus-Hinkler, Suman Singha, Hoyeon Shi, and Matilde Brandt Kreiner
The Cryosphere, 18, 5277–5300, https://doi.org/10.5194/tc-18-5277-2024, https://doi.org/10.5194/tc-18-5277-2024, 2024
Short summary
Short summary
Here, we present ASIP: a new and comprehensive deep-learning-based methodology to retrieve high-resolution sea ice concentration with accompanying well-calibrated uncertainties from satellite-based active and passive microwave observations at a pan-Arctic scale for all seasons. In a comparative study against pan-Arctic ice charts and well-established passive-microwave-based sea ice products, we show that ASIP generalizes well to the pan-Arctic region.
Jean Rabault, Trygve Halsne, Ana Carrasco, Anton Korosov, Joey Voermans, Patrik Bohlinger, Jens Boldingh Debernard, Malte Müller, Øyvind Breivik, Takehiko Nose, Gaute Hope, Fabrice Collard, Sylvain Herlédan, Tsubasa Kodaira, Nick Hughes, Qin Zhang, Kai Haakon Christensen, Alexander Babanin, Lars Willas Dreyer, Cyril Palerme, Lotfi Aouf, Konstantinos Christakos, Atle Jensen, Johannes Röhrs, Aleksey Marchenko, Graig Sutherland, Trygve Kvåle Løken, and Takuji Waseda
EGUsphere, https://doi.org/10.48550/arXiv.2401.07619, https://doi.org/10.48550/arXiv.2401.07619, 2024
Short summary
Short summary
We observe strongly modulated waves-in-ice significant wave height using buoys deployed East of Svalbard. We show that these observations likely cannot be explained by wave-current interaction or tide-induced modulation alone. We also demonstrate a strong correlation between the waves height modulation, and the rate of sea ice convergence. Therefore, our data suggest that the rate of sea ice convergence and divergence may modulate wave in ice energy dissipation.
Xinwei Chen, Muhammed Patel, Fernando J. Pena Cantu, Jinman Park, Javier Noa Turnes, Linlin Xu, K. Andrea Scott, and David A. Clausi
The Cryosphere, 18, 1621–1632, https://doi.org/10.5194/tc-18-1621-2024, https://doi.org/10.5194/tc-18-1621-2024, 2024
Short summary
Short summary
This paper introduces an automated sea ice mapping pipeline utilizing a multi-task U-Net architecture. It attained the top score of 86.3 % in the AutoICE challenge. Ablation studies revealed that incorporating brightness temperature data and spatial–temporal information significantly enhanced model accuracy. Accurate sea ice mapping is vital for comprehending the Arctic environment and its global climate effects, underscoring the potential of deep learning.
Anton Korosov, Pierre Rampal, Yue Ying, Einar Ólason, and Timothy Williams
The Cryosphere, 17, 4223–4240, https://doi.org/10.5194/tc-17-4223-2023, https://doi.org/10.5194/tc-17-4223-2023, 2023
Short summary
Short summary
It is possible to compute sea ice motion from satellite observations and detect areas where ice converges (moves together), forms ice ridges or diverges (moves apart) and opens leads. However, it is difficult to predict the exact motion of sea ice and position of ice ridges or leads using numerical models. We propose a new method to initialise a numerical model from satellite observations to improve the accuracy of the forecasted position of leads and ridges for safer navigation.
Andreas Stokholm, Andrzej Kucik, Nicolas Longépé, and Sine Munk Hvidegaard
EGUsphere, https://doi.org/10.5194/egusphere-2023-976, https://doi.org/10.5194/egusphere-2023-976, 2023
Preprint archived
Short summary
Short summary
To navigate in the Arctic, vessels utilise maps of the local sea ice conditions, that are manually drawn daily. This is a time-consuming and labour-intensive task, which we are trying to automate using convolutional neural networks. In this article, we investigate combining the outputs of models trained with different objectives, leveraging the strengths and avoiding weaknesses to create precise and detailed sea ice concentration maps for navigation in the Arctic and climate or weather models.
Heather Regan, Pierre Rampal, Einar Ólason, Guillaume Boutin, and Anton Korosov
The Cryosphere, 17, 1873–1893, https://doi.org/10.5194/tc-17-1873-2023, https://doi.org/10.5194/tc-17-1873-2023, 2023
Short summary
Short summary
Multiyear ice (MYI), sea ice that survives the summer, is more resistant to changes than younger ice in the Arctic, so it is a good indicator of sea ice resilience. We use a model with a new way of tracking MYI to assess the contribution of different processes affecting MYI. We find two important years for MYI decline: 2007, when dynamics are important, and 2012, when melt is important. These affect MYI volume and area in different ways, which is important for the interpretation of observations.
Karina von Schuckmann, Audrey Minière, Flora Gues, Francisco José Cuesta-Valero, Gottfried Kirchengast, Susheel Adusumilli, Fiammetta Straneo, Michaël Ablain, Richard P. Allan, Paul M. Barker, Hugo Beltrami, Alejandro Blazquez, Tim Boyer, Lijing Cheng, John Church, Damien Desbruyeres, Han Dolman, Catia M. Domingues, Almudena García-García, Donata Giglio, John E. Gilson, Maximilian Gorfer, Leopold Haimberger, Maria Z. Hakuba, Stefan Hendricks, Shigeki Hosoda, Gregory C. Johnson, Rachel Killick, Brian King, Nicolas Kolodziejczyk, Anton Korosov, Gerhard Krinner, Mikael Kuusela, Felix W. Landerer, Moritz Langer, Thomas Lavergne, Isobel Lawrence, Yuehua Li, John Lyman, Florence Marti, Ben Marzeion, Michael Mayer, Andrew H. MacDougall, Trevor McDougall, Didier Paolo Monselesan, Jan Nitzbon, Inès Otosaka, Jian Peng, Sarah Purkey, Dean Roemmich, Kanako Sato, Katsunari Sato, Abhishek Savita, Axel Schweiger, Andrew Shepherd, Sonia I. Seneviratne, Leon Simons, Donald A. Slater, Thomas Slater, Andrea K. Steiner, Toshio Suga, Tanguy Szekely, Wim Thiery, Mary-Louise Timmermans, Inne Vanderkelen, Susan E. Wjiffels, Tonghua Wu, and Michael Zemp
Earth Syst. Sci. Data, 15, 1675–1709, https://doi.org/10.5194/essd-15-1675-2023, https://doi.org/10.5194/essd-15-1675-2023, 2023
Short summary
Short summary
Earth's climate is out of energy balance, and this study quantifies how much heat has consequently accumulated over the past decades (ocean: 89 %, land: 6 %, cryosphere: 4 %, atmosphere: 1 %). Since 1971, this accumulated heat reached record values at an increasing pace. The Earth heat inventory provides a comprehensive view on the status and expectation of global warming, and we call for an implementation of this global climate indicator into the Paris Agreement’s Global Stocktake.
Gifty Attiah, Homa Kheyrollah Pour, and K. Andrea Scott
Earth Syst. Sci. Data, 15, 1329–1355, https://doi.org/10.5194/essd-15-1329-2023, https://doi.org/10.5194/essd-15-1329-2023, 2023
Short summary
Short summary
Lake surface temperature (LST) is a significant indicator of climate change and influences local weather and climate. This study developed a LST dataset retrieved from Landsat archives for 535 lakes across the North Slave Region, NWT, Canada. The data consist of individual NetCDF files for all observed days for each lake. The North Slave LST dataset will provide communities, scientists, and stakeholders with the changing spatiotemporal trends of LST for the past 38 years (1984–2021).
Nazanin Asadi, Philippe Lamontagne, Matthew King, Martin Richard, and K. Andrea Scott
The Cryosphere, 16, 3753–3773, https://doi.org/10.5194/tc-16-3753-2022, https://doi.org/10.5194/tc-16-3753-2022, 2022
Short summary
Short summary
Machine learning approaches are deployed to provide accurate daily spatial maps of sea ice presence probability based on ERA5 data as input. Predictions are capable of predicting freeze-up/breakup dates within a 7 d period at specific locations of interest to shipping operators and communities. Forecasts of the proposed method during the breakup season have skills comparing to Climate Normal and sea ice concentration forecasts from a leading subseasonal-to-seasonal forecasting system.
Stefan Kern, Thomas Lavergne, Leif Toudal Pedersen, Rasmus Tage Tonboe, Louisa Bell, Maybritt Meyer, and Luise Zeigermann
The Cryosphere, 16, 349–378, https://doi.org/10.5194/tc-16-349-2022, https://doi.org/10.5194/tc-16-349-2022, 2022
Short summary
Short summary
High-resolution clear-sky optical satellite imagery has rarely been used to evaluate satellite passive microwave sea-ice concentration products beyond case-study level. By comparing 10 such products with sea-ice concentration estimated from > 350 such optical images in both hemispheres, we expand results of earlier evaluation studies for these products. Results stress the need to look beyond precision and accuracy and to discuss the evaluation data’s quality and filters applied in the products.
Timothy Williams, Anton Korosov, Pierre Rampal, and Einar Ólason
The Cryosphere, 15, 3207–3227, https://doi.org/10.5194/tc-15-3207-2021, https://doi.org/10.5194/tc-15-3207-2021, 2021
Short summary
Short summary
neXtSIM (neXt-generation Sea Ice Model) includes a novel and extremely realistic way of modelling sea ice dynamics – i.e. how the sea ice moves and deforms in response to the drag from winds and ocean currents. It has been developed over the last few years for a variety of applications, but this paper represents its first demonstration in a forecast context. We present results for the time period from November 2018 to June 2020 and show that it agrees well with satellite observations.
Marcel Kleinherenbrink, Anton Korosov, Thomas Newman, Andreas Theodosiou, Alexander S. Komarov, Yuanhao Li, Gert Mulder, Pierre Rampal, Julienne Stroeve, and Paco Lopez-Dekker
The Cryosphere, 15, 3101–3118, https://doi.org/10.5194/tc-15-3101-2021, https://doi.org/10.5194/tc-15-3101-2021, 2021
Short summary
Short summary
Harmony is one of the Earth Explorer 10 candidates that has the chance of being selected for launch in 2028. The mission consists of two satellites that fly in formation with Sentinel-1D, which carries a side-looking radar system. By receiving Sentinel-1's signals reflected from the surface, Harmony is able to observe instantaneous elevation and two-dimensional velocity at the surface. As such, Harmony's data allow the retrieval of sea-ice drift and wave spectra in sea-ice-covered regions.
J. Mifdal, N. Longépé, and M. Rußwurm
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-3-2021, 285–293, https://doi.org/10.5194/isprs-annals-V-3-2021-285-2021, https://doi.org/10.5194/isprs-annals-V-3-2021-285-2021, 2021
Rasmus T. Tonboe, Vishnu Nandan, John Yackel, Stefan Kern, Leif Toudal Pedersen, and Julienne Stroeve
The Cryosphere, 15, 1811–1822, https://doi.org/10.5194/tc-15-1811-2021, https://doi.org/10.5194/tc-15-1811-2021, 2021
Short summary
Short summary
A relationship between the Ku-band radar scattering horizon and snow depth is found using a radar scattering model. This relationship has implications for (1) the use of snow climatology in the conversion of satellite radar freeboard into sea ice thickness and (2) the impact of variability in measured snow depth on the derived ice thickness. For both 1 and 2, the impact of using a snow climatology versus the actual snow depth is relatively small.
Cited articles
Baldwin, S.: Compute Canada: Advancing Computational Research, J. Phys. Conf. Ser. 341, 012001, https://doi.org/10.1088/1742-6596/341/1/012001, 2012. a
Bekkers, E., Francois, J. F., and RojasRomagosa, H.: Melting ice Caps and the Economic Impact of Opening the Northern Sea Route, The Economic Journal, 128, 1095–1127, 2017. a
Boulze, H., Korosov, A., and Brajard, J.: Classification of Sea Ice Types in Sentinel-1 SAR Data Using Convolutional Neural Networks, Remote Sens., 12, 2165, https://doi.org/10.3390/rs12132165, 2020. a
Boutin, G., Williams, T., Rampal, P., Olason, E., and Lique, C.: Impact of wave-induced sea ice fragmentation on sea ice dynamics in the MIZ, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-8657, https://doi.org/10.5194/egusphere-egu2020-8657, 2020. a
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: AI4Arctic Sea Ice Challenge Dataset, DTU [data set], https://doi.org/10.11583/DTU.c.6244065.v2, 2022a. a, b, c, d
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: Raw AI4Arctic Sea Ice Challenge Test Dataset, DTU [data set], https://doi.org/10.11583/DTU.21762848.v1, 2022b. a
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: Ready-To-Train AI4Arctic Sea Ice Challenge Test Dataset, DTU [data set], https://doi.org/10.11583/DTU.21762830.v1, 2022c. a
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: Raw AI4Arctic Sea Ice Challenge Dataset, DTU [data set], https://doi.org/10.11583/DTU.21284967.v3, 2023a. a
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: Ready-To-Train AI4Arctic Sea Ice Challenge Dataset, DTU [data set], https://doi.org/10.11583/DTU.21316608.v3, 2023b. a, b, c, d
Chen, S., Shokr, M., Li, X., Ye, Y., Zhang, Z., Hui, F., and Cheng, X.: MYI Floes Identification Based on the Texture and Shape Feature from Dual-Polarized Sentinel-1 Imagery, Remote Sens., 12, 3221, https://doi.org/10.3390/rs12193221, 2020. a
Chen, X., Patel, M., Pena Cantu, F. J., Park, J., Noa Turnes, J., Xu, L., Scott, K. A., and Clausi, D. A.: MMSeaIce: a collection of techniques for improving sea ice mapping with a multi-task model, The Cryosphere, 18, 1621–1632, https://doi.org/10.5194/tc-18-1621-2024, 2024. a. a, b, c
Chen, X., Patel, M., Xu, L., Chen, Y., Scott, K. A., and Clausi, D. A.: Weakly Supervised Learning for Pixel-Level Sea Ice Concentration Extraction Using AI4Arctic Sea Ice Challenge Dataset, IEEE Geosci. Remote Sens. Lett., 21, 1–5, 2024b. a
Cheng, A., Casati, B., Tivy, A., Zagon, T., Lemieux, J.-F., and Tremblay, L. B.: Accuracy and inter-analyst agreement of visually estimated sea ice concentrations in Canadian Ice Service ice charts using single-polarization RADARSAT-2, The Cryosphere, 14, 1289–1310, https://doi.org/10.5194/tc-14-1289-2020, 2020. a, b
Chevalier, G.: GitHub – Vooban/Smoothly-Blend-Image-Patches: Using a U-Net for image segmentation, blending predicted patches smoothly is a must to please the human eye., https://github.com/Vooban/Smoothly-Blend-Image-Patches (last access: 26 July 2024), 2017. a
Cijov, A.: [Training] – Hubmap CoAT, Kaggle, 2022. a
Cooke, C. L. V. and Scott, K. A.: Estimating Sea Ice Concentration From SAR: Training Convolutional Neural Networks With Passive Microwave Data, IEEE T. Geosci. Remote, 57, 4735–4747, 2019. a
Deng, J., Dong, W., Socher, R., Li, L.-J., Kai Li, and Li Fei-Fei: ImageNet: A large-scale hierarchical image database, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, IEEE, https://doi.org/10.1109/cvpr.2009.5206848, 2009. a
Dunkel, E., Swope, J., Towfic, Z., Chien, S., Russell, D., Sauvageau, J., Sheldon, D., Romero-Cañas, J., Espinosa-Aranda, J. L., Buckley, L., Hervas-Martin, E., Fernandez, M., and Knox, C.: Benchmarking Deep Learning Inference of Remote Sensing Imagery on the Qualcomm Snapdragon And Intel Movidius Myriad X Processors Onboard the International Space Station, 5301–5304, IEEE, 2022. a
Guo, W., Itkin, P., Singha, S., Doulgeris, A. P., Johansson, M., and Spreen, G.: Sea ice classification of TerraSAR-X ScanSAR images for the MOSAiC expedition incorporating per-class incidence angle dependency of image texture, The Cryosphere, 17, 1279–1297, https://doi.org/10.5194/tc-17-1279-2023, 2023. a
Heidler, K., Mou, L., and Zhu, X. X.: Seeing the Bigger Picture: Enabling Large Context Windows in Neural Networks by Combining Multiple Zoom Levels, in: 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, IEEE, 3033–3036, https://doi.org/10.1109/IGARSS47720.2021.9554434, 2021. a, b
Hersbach, H., Bell, B., Berrisford, P., Biavati, G., Horányi, A., Muñoz Sabater, J., Nicolas, J., Peubey, C., Radu, R., Rozum, I., Schepers, D., Simmons, A., Soci, C., Dee, D., and Thépaut, J.-N.: ERA5 hourly data on single levels from 1940 to present, Copernicus Climate Change Service Climate Data Store [data set], https://doi.org/10.24381/cds.adbb2d47, 2023. a
Hoyer, L., Dai, D., and Van Gool, L.: DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation, in: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, https://openaccess.thecvf.com/content/CVPR2022/papers/Hoyer_DAFormer_Improving_Network_Architectures_and_Training_Strategies_for_Domain-Adaptive_Semantic_CVPR_2022_paper.pdf (last access: 29 July 2024), 2022. a
IICWG: SIGRID-3: a vector archive format for sea ice charts, https://library.wmo.int/records/item/37171-sigrid-3-a-vector-archive-format-for-sea-ice-charts (last access: 26 July 2024), 2010. a
Jackson, C. R. and Apel, J. R.: Synthetic Aperture Radar Marine User's Manual, National Environmental Satellite, Data, & Information Service, https://repository.library.noaa.gov/view/noaa/1118 (last access: 29 July 2024), 2004. a
Jiang, M., Clausi, D. A., and Xu, L.: Sea-Ice Mapping of RADARSAT-2 Imagery by Integrating Spatial Contexture With Textural Features, IEEE J. Sel. Top. Appl. Earth Obs., 15, 7964–7977, 2022. a
Karvonen, J.: Baltic Sea ice SAR segmentation and classification using modified pulse-coupled neural networks, IEEE T. Geosci. Remote, 42, 1566–1574, 2004. a
Karvonen, J.: A sea ice concentration estimation algorithm utilizing radiometer and SAR data, The Cryosphere, 8, 1639–1650, https://doi.org/10.5194/tc-8-1639-2014, 2014. a
Kasahara, M., Imaoka, K., Kachi, M., Fujii, H., Naoki, K., Maeda, T., Ito, N., Nakagawa, K., and Oki, T.: Status of AMSR2 on GCOM-W1, in: SPIE Proceedings, https://doi.org/10.1117/12.975810, 2012. a
Khaleghian, S., Ullah, H., Kraemer, T., Eltoft, T., and Marinoni, A.: Deep Semisupervised Teacher–Student Model Based on Label Propagation for Sea Ice Classification, IEEE J. Sel. Top. Appl. Earth Obs., 14, 10761–10772, 2021a. a
Khaleghian, S., Ullah, H., Kræmer, T., Hughes, N., Eltoft, T., and Marinoni, A.: Sea Ice Classification of SAR Imagery Based on Convolution Neural Networks, Remote Sens., 13, 1734, https://doi.org/10.3390/rs13091734, 2021b. a, b
Korosov, A., Demchev, D., Miranda, N., Franceschi, N., and Park, J.-W.: Thermal Denoising of Cross-Polarized Sentinel-1 Data in Interferometric and Extra Wide Swath Modes, IEEE T. Geosci. Remote, 60, 1–11, 2022. a
Kortum, K., Singha, S., and Spreen, G.: Robust Multiseasonal Ice Classification From High-Resolution X-Band SAR, IEEE T. Geosci. Remote, 60, 1–12, 2022. a
Koubarakis, M., Stamoulis, G., Bilidas, D., Ioannidis, T., Mandilaras, G., Pantazi, D.-A., Papadakis, G., Vlassov, V., Payberah2, A., Wang, T., Sheikholeslami, S., Hagos, D. H., Bruzzone, L., Paris, C., Weikmann, G., Marinelli, D., Eltoft, T., Marinoni, A., Kræmer, T., Khaleghian, S., Ullah, H., Troumpoukis, A., Kostopoulou, N. P., Konstantopoulos, S., Karkaletsis, V., Dowling, J., Kakantousis, T., Datcu, M., Yao, W., Dumitru, C. O., Appel, F., Migdall, S., Muerth, M., Bach, H., Hughes, N., Everett, A., Kiærbech, A., Pedersen, J. L., Arthurs, D., Fleming, A., and Cziferszky, A.: Artificial Intelligence and Big Data Technologies for Copernicus Data: The ExtremeEarth Project, in: Publications Office of the EU, 9–12, 2021. a
Kucik, A. and Stokholm, A.: AI4SeaIce: Comparing Loss Representations for SAR Sea Ice Concentration Charting, https://ai4earthscience.github.io/iclr-2022-workshop/accepted (last access: 26 July 2024), 2022. a
Kucik, A. and Stokholm, A.: AI4SeaIce: selecting loss functions for automated SAR sea ice concentration charting, Sci. Rep., 13, 5962, https://doi.org/10.1038/s41598-023-32467-x, 2023. a, b, c
Liu, H., Guo, H., and Liu, G.: A Two-Scale Method of Sea Ice Classification Using TerraSAR-X ScanSAR Data During Early Freeze-Up, IEEE J. Sel. Top. Appl. Earth Obs., 14, 10919–10928, 2021. a
Lyu, H., Huang, W., and Mahdianpari, M.: Eastern Arctic Sea Ice Sensing: First Results from the RADARSAT Constellation Mission Data, Remote Sens., 14, 1165, https://doi.org/10.3390/rs14051165, 2022. a
Malmgren-Hansen, D., Pedersen, L. T., Nielsen, A. A., Skriver, H., Saldo, R., Kreiner, M. B., and Buus-Hinkler, J.: ASIP Sea Ice Dataset – Version 1, https://data.dtu.dk/articles/ASIP_Sea_Ice_Dataset_-_version_1/11920416/1 (last access: 26 July 2024), 2020. a
Nagi, A. S., Kumar, D., Sola, D., and Scott, K. A.: RUF: Effective Sea Ice Floe Segmentation Using End-to-End RES-UNET-CRF with Dual Loss, Remote Sens., 13, 2460, https://doi.org/10.3390/rs13132460, 2021. a
OSI SAF: Global Sea Ice Concentration – DMSP, EUMETSAT SAF on Ocean and Sea Ice, https://doi.org/10.15770/EUM_SAF_OSI_NRT_2004, 2017. a
Papers With Code: ImageNet Benchmark (Image Classification), https://paperswithcode.com/sota/image-classification-on-imagenet (last access: 18 May 2024), 2024. a
Perovich, D., Meier, W., Tschudi, M., Hendricks, S., Petty, A. A., Divine, D., Farrell, S., Gerland, S., Haas, C., Kaleschke, L., Pavlova, O., Ricker, R., Tian-Kunze, X., Webster, M., and Wood, K.: Arctic Report Card 2020: Sea Ice, https://repository.library.noaa.gov/view/noaa/27904 (last access: 26 July 2024), 2020. a
Rogers, M. S., Fox, M., Fleming, A., van Zeeland, L., Wilkinson, J., and Hosking, J. S.: Sea ice detection using concurrent multispectral and synthetic aperture radar imagery, Remote Sens. Environ., 305, 114073, https://doi.org/10.1016/j.rse.2024.114073, 2024. a
Ronneberger, O., Fischer, P., and Brox, T.: U-Net: Convolutional networks for biomedical image segmentation, 234–241, Springer International Publishing, Cham, 2015. a
Stokholm, A.: The AutoICE Competition, Youtube [video], https://youtu.be/iuXIeLPyKfg (last access: 26 July 2024), 2023. a
Stokholm, A., Kucik, A., Wulf, T., Buus-Hinkler, J., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M. B.: GitHub – astokholm/AI4ArcticSeaIceChallenge, GitHub [code], https://github.com/astokholm/AI4ArcticSeaIceChallenge, last access: 26 July 2024. a
Torres, R., Snoeij, P., Geudtner, D., Bibby, D., Davidson, M., Attema, E., Potin, P., Rommen, B., Floury, N., Brown, M., Traver, I. N., Deghaye, P., Duesmann, B., Rosich, B., Miranda, N., Bruno, C., L'Abbate, M., Croci, R., Pietropaolo, A., Huchler, M., and Rostan, F.: GMES Sentinel-1 mission, Remote Sens. Environ., 120, 9–24, 2012. a
Tuia, D., Schindler, K., Demir, B., Camps-Valls, G., Zhu, X. X., Kochupillai, M., Džeroski, S., van Rijn, J. N., Hoos, H. H., Frate, F. D., Datcu, M., Quiané-Ruiz, J.-A., Markl, V., Saux, B. L., and Schneider, R.: Artificial intelligence to advance Earth observation: a perspective, arXiv [preprint], https://arxiv.org/abs/2305.08413 (last access: 26 July 2024), 2023. a
Wang, L., Scott, K. A., Xu, L., and Clausi, D. A.: Sea ice concentration estimation during melt from dual-pol SAR scenes using deep convolutional neural networks: A case study, IEEE T. Geosci. Remote, 54, 4524–4533, 2016. a
Wang, L., Scott, K., and Clausi, D.: Sea Ice Concentration Estimation during Freeze-Up from SAR Imagery Using a Convolutional Neural Network, Remote Sens., 9, 408, https://doi.org/10.3390/rs9050408, 2017a. a
Wang, L., Scott, K. A., Clausi, D. A., and Xu, Y.: Ice concentration estimation in the gulf of St. Lawrence using fully convolutional neural network, in: 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), IEEE, https://doi.org/10.1109/igarss.2017.8128124, 2017b. a
Wang, Q., Lohse, J. P., Doulgeris, A. P., and Eltoft, T.: Data Augmentation for SAR Sea Ice and Water Classification Based on Per-Class Backscatter Variation With Incidence Angle, IEEE T. Geosci. Remote, 61, 1–15, 2023. a
Wulf, T., Buus-Hinkler, J., Singha, S., Shi, H., and Kreiner, M. B.: Pan-Arctic Sea Ice Concentration from SAR and Passive Microwave, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2024-178, 2024. a, b, c
Xu, W., Xu, Y., Chang, T., and Tu, Z.: Co-Scale Conv-Attentional Image Transformers, in: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, 2021. a
Zakhvatkina, N., Korosov, A., Muckenhuber, S., Sandven, S., and Babiker, M.: Operational algorithm for ice–water classification on dual-polarized RADARSAT-2 images, The Cryosphere, 11, 33–46, https://doi.org/10.5194/tc-11-33-2017, 2017. a
Short summary
The AutoICE challenge encouraged the development of deep learning models to map multiple aspects of sea ice – the amount of sea ice in an area and the age and ice floe size – using multiple sources of satellite and weather data across the Canadian and Greenlandic Arctic. Professionally drawn operational sea ice charts were used as a reference. A total of 179 students and sea ice and AI specialists participated and produced maps in broad agreement with the sea ice charts.
The AutoICE challenge encouraged the development of deep learning models to map multiple aspects...
Special issue