Articles | Volume 18, issue 4
https://doi.org/10.5194/tc-18-1621-2024
© 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-18-1621-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
08 Apr 2024
Research article |
| 08 Apr 2024
| 08 Apr 2024
MMSeaIce: a collection of techniques for improving sea ice mapping with a multi-task model
Xinwei Chen
School of Marine Science and Engineering, South China University of Technology, Guangzhou, China
Muhammed Patel
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
Fernando J. Pena Cantu
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
Jinman Park
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
Javier Noa Turnes
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
Linlin Xu
CORRESPONDING AUTHOR
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
K. Andrea Scott
Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, ON, Canada
David A. Clausi
Vision and Image Processing Lab, Department of System Design Engineering, University of Waterloo, Waterloo, ON, Canada
Related authors
Andreas R. Stokholm, Jørgen Buus-Hinkler, Tore Wulf, Anton Korosov, Roberto Saldo, Leif T. Pedersen, David Arthurs, Ionut Dragan, Iacopo Modica, Juan Pedro, Annekatrien Debien, Xinwei Chen, Muhammed Patel, Fernando J. P. Cantu, Javier N. Turnes, Jinman Park, Linlin Xu, Andrea K. Scott, David A. Clausi, Yuan Fang, Mingzhe Jiang, Saeid Taleghanidoozdoozan, Neil C. Brubacher, Armina Soleymani, Zacharie Gousseau, Michał Smaczny, Patryk Kowalski, Jacek Komorowski, David Rijlaarsdam, Jan N. van Rijn, Jens Jakobsen, Martin S. J. Rogers, Nick Hughes, Tom Zagon, Rune Solberg, Nicolas Longépé, and Matilde B. Kreiner
EGUsphere, https://doi.org/10.5194/egusphere-2023-2648, https://doi.org/10.5194/egusphere-2023-2648, 2023
Short summary
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, the age and ice floe size using multiple sources of satellite and weather data across the Canadian and Greenlandic Arctic. To judge the solution, professionally drawn operational sea ice charts were used as reference. 179 students and sea ice and AI specialists participated and produced maps in broad agreement with the sea ice charts.
Andreas R. Stokholm, Jørgen Buus-Hinkler, Tore Wulf, Anton Korosov, Roberto Saldo, Leif T. Pedersen, David Arthurs, Ionut Dragan, Iacopo Modica, Juan Pedro, Annekatrien Debien, Xinwei Chen, Muhammed Patel, Fernando J. P. Cantu, Javier N. Turnes, Jinman Park, Linlin Xu, Andrea K. Scott, David A. Clausi, Yuan Fang, Mingzhe Jiang, Saeid Taleghanidoozdoozan, Neil C. Brubacher, Armina Soleymani, Zacharie Gousseau, Michał Smaczny, Patryk Kowalski, Jacek Komorowski, David Rijlaarsdam, Jan N. van Rijn, Jens Jakobsen, Martin S. J. Rogers, Nick Hughes, Tom Zagon, Rune Solberg, Nicolas Longépé, and Matilde B. Kreiner
EGUsphere, https://doi.org/10.5194/egusphere-2023-2648, https://doi.org/10.5194/egusphere-2023-2648, 2023
Short summary
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, the age and ice floe size using multiple sources of satellite and weather data across the Canadian and Greenlandic Arctic. To judge the solution, professionally drawn operational sea ice charts were used as reference. 179 students and sea ice and AI specialists participated and produced maps in broad agreement with the sea ice charts.
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.
Y. Fang, L. Xu, J. Peng, H. Wang, A. Wong, and D. A. Clausi
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3, 335–340, https://doi.org/10.5194/isprs-archives-XLII-3-335-2018, https://doi.org/10.5194/isprs-archives-XLII-3-335-2018, 2018
Related subject area
Discipline: Sea ice | Subject: Remote Sensing
Sea ice transport and replenishment across and within the Canadian Arctic Archipelago, 2016–2022
SAR deep learning sea ice retrieval trained with airborne laser scanner measurements from the MOSAiC expedition
Lead fractions from SAR-derived sea ice divergence during MOSAiC
Ice floe segmentation and floe size distribution in airborne and high-resolution optical satellite images: towards an automated labelling deep learning approach
New estimates of pan-Arctic sea ice–atmosphere neutral drag coefficients from ICESat-2 elevation data
Relevance of warm air intrusions for Arctic satellite sea ice concentration time series
Observing the evolution of summer melt on multiyear sea ice with ICESat-2 and Sentinel-2
Estimating differential penetration of green (532 nm) laser light over sea ice with NASA’s Airborne Topographic Mapper: observations and models
Spaceborne thermal infrared observations of Arctic sea ice leads at 30 m resolution
Wind redistribution of snow impacts the Ka- and Ku-band radar signatures of Arctic sea ice
First observations of sea ice flexural–gravity waves with ground-based radar interferometry in Utqiaġvik, Alaska
Feasibility of retrieving Arctic sea ice thickness from the Chinese HY-2B Ku-band radar altimeter
Sea ice classification of TerraSAR-X ScanSAR images for the MOSAiC expedition incorporating per-class incidence angle dependency of image texture
Aerial observations of sea ice breakup by ship waves
Monitoring Arctic thin ice: a comparison between CryoSat-2 SAR altimetry data and MODIS thermal-infrared imagery
The effects of surface roughness on the calculated, spectral, conical–conical reflectance factor as an alternative to the bidirectional reflectance distribution function of bare sea ice
Inter-comparison and evaluation of Arctic sea ice type products
Estimating the uncertainty of sea-ice area and sea-ice extent from satellite retrievals
A simple model for daily basin-wide thermodynamic sea ice thickness growth retrieval
Ice ridge density signatures in high-resolution SAR images
Rain on snow (ROS) understudied in sea ice remote sensing: a multi-sensor analysis of ROS during MOSAiC (Multidisciplinary drifting Observatory for the Study of Arctic Climate)
Quantifying the effects of background concentrations of crude oil pollution on sea ice albedo
Characterizing the sea-ice floe size distribution in the Canada Basin from high-resolution optical satellite imagery
Generating large-scale sea ice motion from Sentinel-1 and the RADARSAT Constellation Mission using the Environment and Climate Change Canada automated sea ice tracking system
Rotational drift in Antarctic sea ice: pronounced cyclonic features and differences between data products
Satellite passive microwave sea-ice concentration data set intercomparison using Landsat data
Cross-platform classification of level and deformed sea ice considering per-class incident angle dependency of backscatter intensity
Advances in altimetric snow depth estimates using bi-frequency SARAL and CryoSat-2 Ka–Ku measurements
Antarctic snow-covered sea ice topography derivation from TanDEM-X using polarimetric SAR interferometry
Impacts of snow data and processing methods on the interpretation of long-term changes in Baffin Bay early spring sea ice thickness
A lead-width distribution for Antarctic sea ice: a case study for the Weddell Sea with high-resolution Sentinel-2 images
Satellite altimetry detection of ice-shelf-influenced fast ice
MOSAiC drift expedition from October 2019 to July 2020: sea ice conditions from space and comparison with previous years
Towards a swath-to-swath sea-ice drift product for the Copernicus Imaging Microwave Radiometer mission
Spaceborne infrared imagery for early detection of Weddell Polynya opening
Estimating instantaneous sea-ice dynamics from space using the bi-static radar measurements of Earth Explorer 10 candidate Harmony
Estimating subpixel turbulent heat flux over leads from MODIS thermal infrared imagery with deep learning
An improved sea ice detection algorithm using MODIS: application as a new European sea ice extent indicator
Faster decline and higher variability in the sea ice thickness of the marginal Arctic seas when accounting for dynamic snow cover
Estimation of degree of sea ice ridging in the Bay of Bothnia based on geolocated photon heights from ICESat-2
Linking sea ice deformation to ice thickness redistribution using high-resolution satellite and airborne observations
Simulated Ka- and Ku-band radar altimeter height and freeboard estimation on snow-covered Arctic sea ice
Improved machine-learning-based open-water–sea-ice–cloud discrimination over wintertime Antarctic sea ice using MODIS thermal-infrared imagery
Spring melt pond fraction in the Canadian Arctic Archipelago predicted from RADARSAT-2
Simultaneous estimation of wintertime sea ice thickness and snow depth from space-borne freeboard measurements
Observations of sea ice melt from Operation IceBridge imagery
Estimating statistical errors in retrievals of ice velocity and deformation parameters from satellite images and buoy arrays
Brief Communication: Mesoscale and submesoscale dynamics in the marginal ice zone from sequential synthetic aperture radar observations
Classification of sea ice types in Sentinel-1 synthetic aperture radar images
A linear model to derive melt pond depth on Arctic sea ice from hyperspectral data
Stephen E. L. Howell, David G. Babb, Jack C. Landy, Isolde A. Glissenaar, Kaitlin McNeil, Benoit Montpetit, and Mike Brady
The Cryosphere, 18, 2321–2333, https://doi.org/10.5194/tc-18-2321-2024, https://doi.org/10.5194/tc-18-2321-2024, 2024
Short summary
Short summary
The CAA serves as both a source and a sink for sea ice from the Arctic Ocean, while also exporting sea ice into Baffin Bay. It is also an important region with respect to navigating the Northwest Passage. Here, we quantify sea ice transport and replenishment across and within the CAA from 2016 to 2022. We also provide the first estimates of the ice area and volume flux within the CAA from the Queen Elizabeth Islands to Parry Channel, which spans the central region of the Northwest Passage.
Karl Kortum, Suman Singha, Gunnar Spreen, Nils Hutter, Arttu Jutila, and Christian Haas
The Cryosphere, 18, 2207–2222, https://doi.org/10.5194/tc-18-2207-2024, https://doi.org/10.5194/tc-18-2207-2024, 2024
Short summary
Short summary
A dataset of 20 radar satellite acquisitions and near-simultaneous helicopter-based surveys of the ice topography during the MOSAiC expedition is constructed and used to train a variety of deep learning algorithms. The results give realistic insights into the accuracy of retrieval of measured ice classes using modern deep learning models. The models able to learn from the spatial distribution of the measured sea ice classes are shown to have a clear advantage over those that cannot.
Luisa von Albedyll, Stefan Hendricks, Nils Hutter, Dmitrii Murashkin, Lars Kaleschke, Sascha Willmes, Linda Thielke, Xiangshan Tian-Kunze, Gunnar Spreen, and Christian Haas
The Cryosphere, 18, 1259–1285, https://doi.org/10.5194/tc-18-1259-2024, https://doi.org/10.5194/tc-18-1259-2024, 2024
Short summary
Short summary
Leads (openings in sea ice cover) are created by sea ice dynamics. Because they are important for many processes in the Arctic winter climate, we aim to detect them with satellites. We present two new techniques to detect lead widths of a few hundred meters at high spatial resolution (700 m) and independent of clouds or sun illumination. We use the MOSAiC drift 2019–2020 in the Arctic for our case study and compare our new products to other existing lead products.
Qin Zhang and Nick Hughes
The Cryosphere, 17, 5519–5537, https://doi.org/10.5194/tc-17-5519-2023, https://doi.org/10.5194/tc-17-5519-2023, 2023
Short summary
Short summary
To alleviate tedious manual image annotations for training deep learning (DL) models in floe instance segmentation, we employ a classical image processing technique to automatically label floes in images. We then apply a DL semantic method for fast and adaptive floe instance segmentation from high-resolution airborne and satellite images. A post-processing algorithm is also proposed to refine the segmentation and further to derive acceptable floe size distributions at local and global scales.
Alexander Mchedlishvili, Christof Lüpkes, Alek Petty, Michel Tsamados, and Gunnar Spreen
The Cryosphere, 17, 4103–4131, https://doi.org/10.5194/tc-17-4103-2023, https://doi.org/10.5194/tc-17-4103-2023, 2023
Short summary
Short summary
In this study we looked at sea ice–atmosphere drag coefficients, quantities that help with characterizing the friction between the atmosphere and sea ice, and vice versa. Using ICESat-2, a laser altimeter that measures elevation differences by timing how long it takes for photons it sends out to return to itself, we could map the roughness, i.e., how uneven the surface is. From roughness we then estimate drag force, the frictional force between sea ice and the atmosphere, across the Arctic.
Philip Rostosky and Gunnar Spreen
The Cryosphere, 17, 3867–3881, https://doi.org/10.5194/tc-17-3867-2023, https://doi.org/10.5194/tc-17-3867-2023, 2023
Short summary
Short summary
During winter, storms entering the Arctic region can bring warm air into the cold environment. Strong increases in air temperature modify the characteristics of the Arctic snow and ice cover. The Arctic sea ice cover can be monitored by satellites observing the natural emission of the Earth's surface. In this study, we show that during warm air intrusions the change in the snow characteristics influences the satellite-derived sea ice cover, leading to a false reduction of the estimated ice area.
Ellen M. Buckley, Sinéad L. Farrell, Ute C. Herzfeld, Melinda A. Webster, Thomas Trantow, Oliwia N. Baney, Kyle A. Duncan, Huilin Han, and Matthew Lawson
The Cryosphere, 17, 3695–3719, https://doi.org/10.5194/tc-17-3695-2023, https://doi.org/10.5194/tc-17-3695-2023, 2023
Short summary
Short summary
In this study, we use satellite observations to investigate the evolution of melt ponds on the Arctic sea ice surface. We derive melt pond depth from ICESat-2 measurements of the pond surface and bathymetry and melt pond fraction (MPF) from the classification of Sentinel-2 imagery. MPF increases to a peak of 16 % in late June and then decreases, while depth increases steadily. This work demonstrates the ability to track evolving melt conditions in three dimensions throughout the summer.
Michael Studinger, Benjamin E. Smith, Nathan Kurtz, Alek Petty, Tyler Sutterley, and Rachel Tilling
The Cryosphere Discuss., https://doi.org/10.5194/tc-2023-126, https://doi.org/10.5194/tc-2023-126, 2023
Revised manuscript accepted for TC
Short summary
Short summary
We use green lidar data and natural color imagery over sea ice to quantify elevation biases potentially impacting estimates of change in ice thickness of the polar regions. We complement our analysis using a model of scattering of light in snow and ice that predicts the shape of lidar waveforms reflecting from snow and ice surfaces based on the shape of the transmitted pulse. We find that biased elevations exist in airborne and spaceborne data products from green lidars.
Yujia Qiu, Xiao-Ming Li, and Huadong Guo
The Cryosphere, 17, 2829–2849, https://doi.org/10.5194/tc-17-2829-2023, https://doi.org/10.5194/tc-17-2829-2023, 2023
Short summary
Short summary
Spaceborne thermal infrared sensors with kilometer-scale resolution cannot support adequate parameterization of Arctic leads. For the first time, we applied the 30 m resolution data from the Thermal Infrared Spectrometer (TIS) on the emerging SDGSAT-1 to detect Arctic leads. Validation with Sentinel-2 data shows high accuracy for the three TIS bands. Compared to MODIS, the TIS presents more narrow leads, demonstrating its great potential for observing previously unresolvable Arctic leads.
Vishnu Nandan, Rosemary Willatt, Robbie Mallett, Julienne Stroeve, Torsten Geldsetzer, Randall Scharien, Rasmus Tonboe, John Yackel, Jack Landy, David Clemens-Sewall, Arttu Jutila, David N. Wagner, Daniela Krampe, Marcus Huntemann, Mallik Mahmud, David Jensen, Thomas Newman, Stefan Hendricks, Gunnar Spreen, Amy Macfarlane, Martin Schneebeli, James Mead, Robert Ricker, Michael Gallagher, Claude Duguay, Ian Raphael, Chris Polashenski, Michel Tsamados, Ilkka Matero, and Mario Hoppmann
The Cryosphere, 17, 2211–2229, https://doi.org/10.5194/tc-17-2211-2023, https://doi.org/10.5194/tc-17-2211-2023, 2023
Short summary
Short summary
We show that wind redistributes snow on Arctic sea ice, and Ka- and Ku-band radar measurements detect both newly deposited snow and buried snow layers that can affect the accuracy of snow depth estimates on sea ice. Radar, laser, meteorological, and snow data were collected during the MOSAiC expedition. With frequent occurrence of storms in the Arctic, our results show that
wind-redistributed snow needs to be accounted for to improve snow depth estimates on sea ice from satellite radars.
Dyre Oliver Dammann, Mark A. Johnson, Andrew R. Mahoney, and Emily R. Fedders
The Cryosphere, 17, 1609–1622, https://doi.org/10.5194/tc-17-1609-2023, https://doi.org/10.5194/tc-17-1609-2023, 2023
Short summary
Short summary
We investigate the GAMMA Portable Radar Interferometer (GPRI) as a tool for evaluating flexural–gravity waves in sea ice in near real time. With a GPRI mounted on grounded ice near Utqiaġvik, Alaska, we identify 20–50 s infragravity waves in landfast ice with ~1 mm amplitude during 23–24 April 2021. Observed wave speed and periods compare well with modeled wave propagation and on-ice accelerometers, confirming the ability to track propagation and properties of waves over hundreds of meters.
Zhaoqing Dong, Lijian Shi, Mingsen Lin, Yongjun Jia, Tao Zeng, and Suhui Wu
The Cryosphere, 17, 1389–1410, https://doi.org/10.5194/tc-17-1389-2023, https://doi.org/10.5194/tc-17-1389-2023, 2023
Short summary
Short summary
We try to explore the application of SGDR data in polar sea ice thickness. Through this study, we find that it seems difficult to obtain reasonable results by using conventional methods. So we use the 15 lowest points per 25 km to estimate SSHA to retrieve more reasonable Arctic radar freeboard and thickness. This study also provides reference for reprocessing L1 data. We will release products that are more reasonable and suitable for polar sea ice thickness retrieval to better evaluate HY-2B.
Wenkai Guo, Polona Itkin, Suman Singha, Anthony P. Doulgeris, Malin Johansson, and Gunnar Spreen
The Cryosphere, 17, 1279–1297, https://doi.org/10.5194/tc-17-1279-2023, https://doi.org/10.5194/tc-17-1279-2023, 2023
Short summary
Short summary
Sea ice maps are produced to cover the MOSAiC Arctic expedition (2019–2020) and divide sea ice into scientifically meaningful classes. We use a high-resolution X-band synthetic aperture radar dataset and show how image brightness and texture systematically vary across the images. We use an algorithm that reliably corrects this effect and achieve good results, as evaluated by comparisons to ground observations and other studies. The sea ice maps are useful as a basis for future MOSAiC studies.
Elie Dumas-Lefebvre and Dany Dumont
The Cryosphere, 17, 827–842, https://doi.org/10.5194/tc-17-827-2023, https://doi.org/10.5194/tc-17-827-2023, 2023
Short summary
Short summary
By changing the shape of ice floes, wave-induced sea ice breakup dramatically affects the large-scale dynamics of sea ice. As this process is also the trigger of multiple others, it was deemed relevant to study how breakup itself affects the ice floe size distribution. To do so, a ship sailed close to ice floes, and the breakup that it generated was recorded with a drone. The obtained data shed light on the underlying physics of wave-induced sea ice breakup.
Felix L. Müller, Stephan Paul, Stefan Hendricks, and Denise Dettmering
The Cryosphere, 17, 809–825, https://doi.org/10.5194/tc-17-809-2023, https://doi.org/10.5194/tc-17-809-2023, 2023
Short summary
Short summary
Thinning sea ice has significant impacts on the energy exchange between the atmosphere and the ocean. In this study we present visual and quantitative comparisons of thin-ice detections obtained from classified Cryosat-2 radar reflections and thin-ice-thickness estimates derived from MODIS thermal-infrared imagery. In addition to good comparability, the results of the study indicate the potential for a deeper understanding of sea ice in the polar seas and improved processing of altimeter data.
Maxim L. Lamare, John D. Hedley, and Martin D. King
The Cryosphere, 17, 737–751, https://doi.org/10.5194/tc-17-737-2023, https://doi.org/10.5194/tc-17-737-2023, 2023
Short summary
Short summary
The reflectivity of sea ice is crucial for modern climate change and for monitoring sea ice from satellites. The reflectivity depends on the angle at which the ice is viewed and the angle illuminated. The directional reflectivity is calculated as a function of viewing angle, illuminating angle, thickness, wavelength and surface roughness. Roughness cannot be considered independent of thickness, illumination angle and the wavelength. Remote sensors will use the data to image sea ice from space.
Yufang Ye, Yanbing Luo, Yan Sun, Mohammed Shokr, Signe Aaboe, Fanny Girard-Ardhuin, Fengming Hui, Xiao Cheng, and Zhuoqi Chen
The Cryosphere, 17, 279–308, https://doi.org/10.5194/tc-17-279-2023, https://doi.org/10.5194/tc-17-279-2023, 2023
Short summary
Short summary
Arctic sea ice type (SITY) variation is a sensitive indicator of climate change. This study gives a systematic inter-comparison and evaluation of eight SITY products. Main results include differences in SITY products being significant, with average Arctic multiyear ice extent up to 1.8×106 km2; Ku-band scatterometer SITY products generally performing better; and factors such as satellite inputs, classification methods, training datasets and post-processing highly impacting their performance.
Andreas Wernecke, Dirk Notz, Stefan Kern, and Thomas Lavergne
EGUsphere, https://doi.org/10.5194/egusphere-2022-1189, https://doi.org/10.5194/egusphere-2022-1189, 2022
Short summary
Short summary
The total area of Arctic sea ice (SIA) is routinely monitored with the help of satellite measurements, which is an important climate indicator. Uncertainties in observations of sea-ice concentration (SIC) cancel out when averaging but the degree to which this is happening has been unclear. We calculate the uncertainty in SIA based on uncertainties in SIC and find that daily SIA estimates have uncertainties of about 300 000 km2. The 2002 to 2017 September trend is 105 000 ± 9 000 km2 per year.
James Anheuser, Yinghui Liu, and Jeffrey R. Key
The Cryosphere, 16, 4403–4421, https://doi.org/10.5194/tc-16-4403-2022, https://doi.org/10.5194/tc-16-4403-2022, 2022
Short summary
Short summary
A prominent part of the polar climate system is sea ice, a better understanding of which would lead to better understanding Earth's climate. Newly published methods for observing the temperature of sea ice have made possible a new method for estimating daily sea ice thickness growth from space using an energy balance. The method compares well with existing sea ice thickness observations.
Mikko Lensu and Markku Similä
The Cryosphere, 16, 4363–4377, https://doi.org/10.5194/tc-16-4363-2022, https://doi.org/10.5194/tc-16-4363-2022, 2022
Short summary
Short summary
Ice ridges form a compressing ice cover. From above they appear as walls of up to few metres in height and extend even kilometres across the ice. Below they may reach tens of metres under the sea surface. Ridges need to be observed for the purposes of ice forecasting and ice information production. This relies mostly on ridging signatures discernible in radar satellite (SAR) images. New methods to quantify ridging from SAR have been developed and are shown to agree with field observations.
Julienne Stroeve, Vishnu Nandan, Rosemary Willatt, Ruzica Dadic, Philip Rostosky, Michael Gallagher, Robbie Mallett, Andrew Barrett, Stefan Hendricks, Rasmus Tonboe, Michelle McCrystall, Mark Serreze, Linda Thielke, Gunnar Spreen, Thomas Newman, John Yackel, Robert Ricker, Michel Tsamados, Amy Macfarlane, Henna-Reetta Hannula, and Martin Schneebeli
The Cryosphere, 16, 4223–4250, https://doi.org/10.5194/tc-16-4223-2022, https://doi.org/10.5194/tc-16-4223-2022, 2022
Short summary
Short summary
Impacts of rain on snow (ROS) on satellite-retrieved sea ice variables remain to be fully understood. This study evaluates the impacts of ROS over sea ice on active and passive microwave data collected during the 2019–20 MOSAiC expedition. Rainfall and subsequent refreezing of the snowpack significantly altered emitted and backscattered radar energy, laying important groundwork for understanding their impacts on operational satellite retrievals of various sea ice geophysical variables.
Benjamin Heikki Redmond Roche and Martin D. King
The Cryosphere, 16, 3949–3970, https://doi.org/10.5194/tc-16-3949-2022, https://doi.org/10.5194/tc-16-3949-2022, 2022
Short summary
Short summary
Sea ice is bright, playing an important role in reflecting incoming solar radiation. The reflectivity of sea ice is affected by the presence of pollutants, such as crude oil, even at low concentrations. Modelling how the brightness of three types of sea ice is affected by increasing concentrations of crude oils shows that the type of oil, the type of ice, the thickness of the ice, and the size of the oil droplets are important factors. This shows that sea ice is vulnerable to oil pollution.
Alexis Anne Denton and Mary-Louise Timmermans
The Cryosphere, 16, 1563–1578, https://doi.org/10.5194/tc-16-1563-2022, https://doi.org/10.5194/tc-16-1563-2022, 2022
Short summary
Short summary
Arctic sea ice has a distribution of ice sizes that provides insight into the physics of the ice. We examine this distribution from satellite imagery from 1999 to 2014 in the Canada Basin. We find that it appears as a power law whose power becomes less negative with increasing ice concentrations and has a seasonality tied to that of ice concentration. Results suggest ice concentration be considered in models of this distribution and are important for understanding sea ice in a warming Arctic.
Stephen E. L. Howell, Mike Brady, and Alexander S. Komarov
The Cryosphere, 16, 1125–1139, https://doi.org/10.5194/tc-16-1125-2022, https://doi.org/10.5194/tc-16-1125-2022, 2022
Short summary
Short summary
We describe, apply, and validate the Environment and Climate Change Canada automated sea ice tracking system (ECCC-ASITS) that routinely generates large-scale sea ice motion (SIM) over the pan-Arctic domain using synthetic aperture radar (SAR) images. The ECCC-ASITS was applied to the incoming image streams of Sentinel-1AB and the RADARSAT Constellation Mission from March 2020 to October 2021 using a total of 135 471 SAR images and generated new SIM datasets (i.e., 7 d 25 km and 3 d 6.25 km).
Wayne de Jager and Marcello Vichi
The Cryosphere, 16, 925–940, https://doi.org/10.5194/tc-16-925-2022, https://doi.org/10.5194/tc-16-925-2022, 2022
Short summary
Short summary
Ice motion can be used to better understand how weather and climate change affect the ice. Antarctic sea ice extent has shown large variability over the observed period, and dynamical features may also have changed. Our method allows for the quantification of rotational motion caused by wind and how this may have changed with time. Cyclonic motion dominates the Atlantic sector, particularly from 2015 onwards, while anticyclonic motion has remained comparatively small and unchanged.
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.
Wenkai Guo, Polona Itkin, Johannes Lohse, Malin Johansson, and Anthony Paul Doulgeris
The Cryosphere, 16, 237–257, https://doi.org/10.5194/tc-16-237-2022, https://doi.org/10.5194/tc-16-237-2022, 2022
Short summary
Short summary
This study uses radar satellite data categorized into different sea ice types to detect ice deformation, which is significant for climate science and ship navigation. For this, we examine radar signal differences of sea ice between two similar satellite sensors and show an optimal way to apply categorization methods across sensors, so more data can be used for this purpose. This study provides a basis for future reliable and constant detection of ice deformation remotely through satellite data.
Florent Garnier, Sara Fleury, Gilles Garric, Jérôme Bouffard, Michel Tsamados, Antoine Laforge, Marion Bocquet, Renée Mie Fredensborg Hansen, and Frédérique Remy
The Cryosphere, 15, 5483–5512, https://doi.org/10.5194/tc-15-5483-2021, https://doi.org/10.5194/tc-15-5483-2021, 2021
Short summary
Short summary
Snow depth data are essential to monitor the impacts of climate change on sea ice volume variations and their impacts on the climate system. For that purpose, we present and assess the altimetric snow depth product, computed in both hemispheres from CryoSat-2 and SARAL satellite data. The use of these data instead of the common climatology reduces the sea ice thickness by about 30 cm over the 2013–2019 period. These data are also crucial to argue for the launch of the CRISTAL satellite mission.
Lanqing Huang, Georg Fischer, and Irena Hajnsek
The Cryosphere, 15, 5323–5344, https://doi.org/10.5194/tc-15-5323-2021, https://doi.org/10.5194/tc-15-5323-2021, 2021
Short summary
Short summary
This study shows an elevation difference between the radar interferometric measurements and the optical measurements from a coordinated campaign over the snow-covered deformed sea ice in the western Weddell Sea, Antarctica. The objective is to correct the penetration bias of microwaves and to generate a precise sea ice topographic map, including the snow depth on top. Excellent performance for sea ice topographic retrieval is achieved with the proposed model and the developed retrieval scheme.
Isolde A. Glissenaar, Jack C. Landy, Alek A. Petty, Nathan T. Kurtz, and Julienne C. Stroeve
The Cryosphere, 15, 4909–4927, https://doi.org/10.5194/tc-15-4909-2021, https://doi.org/10.5194/tc-15-4909-2021, 2021
Short summary
Short summary
Scientists can estimate sea ice thickness using satellites that measure surface height. To determine the sea ice thickness, we also need to know the snow depth and density. This paper shows that the chosen snow depth product has a considerable impact on the findings of sea ice thickness state and trends in Baffin Bay, showing mean thinning with some snow depth products and mean thickening with others. This shows that it is important to better understand and monitor snow depth on sea ice.
Marek Muchow, Amelie U. Schmitt, and Lars Kaleschke
The Cryosphere, 15, 4527–4537, https://doi.org/10.5194/tc-15-4527-2021, https://doi.org/10.5194/tc-15-4527-2021, 2021
Short summary
Short summary
Linear-like openings in sea ice, also called leads, occur with widths from meters to kilometers. We use satellite images from Sentinel-2 with a resolution of 10 m to identify leads and measure their widths. With that we investigate the frequency of lead widths using two different statistical methods, since other studies have shown a dependency of heat exchange on the lead width. We are the first to address the sea-ice lead-width distribution in the Weddell Sea, Antarctica.
Gemma M. Brett, Daniel Price, Wolfgang Rack, and Patricia J. Langhorne
The Cryosphere, 15, 4099–4115, https://doi.org/10.5194/tc-15-4099-2021, https://doi.org/10.5194/tc-15-4099-2021, 2021
Short summary
Short summary
Ice shelf meltwater in the surface ocean affects sea ice formation, causing it to be thicker and, in particular conditions, to have a loose mass of platelet ice crystals called a sub‐ice platelet layer beneath. This causes the sea ice freeboard to stand higher above sea level. In this study, we demonstrate for the first time that the signature of ice shelf meltwater in the surface ocean manifesting as higher sea ice freeboard in McMurdo Sound is detectable from space using satellite technology.
Thomas Krumpen, Luisa von Albedyll, Helge F. Goessling, Stefan Hendricks, Bennet Juhls, Gunnar Spreen, Sascha Willmes, H. Jakob Belter, Klaus Dethloff, Christian Haas, Lars Kaleschke, Christian Katlein, Xiangshan Tian-Kunze, Robert Ricker, Philip Rostosky, Janna Rückert, Suman Singha, and Julia Sokolova
The Cryosphere, 15, 3897–3920, https://doi.org/10.5194/tc-15-3897-2021, https://doi.org/10.5194/tc-15-3897-2021, 2021
Short summary
Short summary
We use satellite data records collected along the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) drift to categorize ice conditions that shaped and characterized the floe and surroundings during the expedition. A comparison with previous years is made whenever possible. The aim of this analysis is to provide a basis and reference for subsequent research in the six main research areas of atmosphere, ocean, sea ice, biogeochemistry, remote sensing and ecology.
Thomas Lavergne, Montserrat Piñol Solé, Emily Down, and Craig Donlon
The Cryosphere, 15, 3681–3698, https://doi.org/10.5194/tc-15-3681-2021, https://doi.org/10.5194/tc-15-3681-2021, 2021
Short summary
Short summary
Pushed by winds and ocean currents, polar sea ice is on the move. We use passive microwave satellites to observe this motion. The images from their orbits are often put together into daily images before motion is measured. In our study, we measure motion from the individual orbits directly and not from the daily images. We obtain many more motion vectors, and they are more accurate. This can be used for current and future satellites, e.g. the Copernicus Imaging Microwave Radiometer (CIMR).
Céline Heuzé, Lu Zhou, Martin Mohrmann, and Adriano Lemos
The Cryosphere, 15, 3401–3421, https://doi.org/10.5194/tc-15-3401-2021, https://doi.org/10.5194/tc-15-3401-2021, 2021
Short summary
Short summary
For navigation or science planning, knowing when sea ice will open in advance is a prerequisite. Yet, to date, routine spaceborne microwave observations of sea ice are unable to do so. We present the first method based on spaceborne infrared that can forecast an opening several days ahead. We develop it specifically for the Weddell Polynya, a large hole in the Antarctic winter ice cover that unexpectedly re-opened for the first time in 40 years in 2016, and determine why the polynya opened.
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.
Zhixiang Yin, Xiaodong Li, Yong Ge, Cheng Shang, Xinyan Li, Yun Du, and Feng Ling
The Cryosphere, 15, 2835–2856, https://doi.org/10.5194/tc-15-2835-2021, https://doi.org/10.5194/tc-15-2835-2021, 2021
Short summary
Short summary
MODIS thermal infrared (TIR) imagery provides promising data to study the rapid variations in the Arctic turbulent heat flux (THF). The accuracy of estimated THF, however, is low (especially for small leads) due to the coarse resolution of the MODIS TIR data. We train a deep neural network to enhance the spatial resolution of estimated THF over leads from MODIS TIR imagery. The method is found to be effective and can generate a result which is close to that derived from Landsat-8 TIR imagery.
Joan Antoni Parera-Portell, Raquel Ubach, and Charles Gignac
The Cryosphere, 15, 2803–2818, https://doi.org/10.5194/tc-15-2803-2021, https://doi.org/10.5194/tc-15-2803-2021, 2021
Short summary
Short summary
We describe a new method to map sea ice and water at 500 m resolution using data acquired by the MODIS sensors. The strength of this method is that it achieves high-accuracy results and is capable of attenuating unwanted resolution-breaking effects caused by cloud masking. Our resulting March and September monthly aggregates reflect the loss of sea ice in the European Arctic during the 2000–2019 period and show the algorithm's usefulness as a sea ice monitoring tool.
Robbie D. C. Mallett, Julienne C. Stroeve, Michel Tsamados, Jack C. Landy, Rosemary Willatt, Vishnu Nandan, and Glen E. Liston
The Cryosphere, 15, 2429–2450, https://doi.org/10.5194/tc-15-2429-2021, https://doi.org/10.5194/tc-15-2429-2021, 2021
Short summary
Short summary
We re-estimate pan-Arctic sea ice thickness (SIT) values by combining data from the Envisat and CryoSat-2 missions with data from a new, reanalysis-driven snow model. Because a decreasing amount of ice is being hidden below the waterline by the weight of overlying snow, we argue that SIT may be declining faster than previously calculated in some regions. Because the snow product varies from year to year, our new SIT calculations also display much more year-to-year variability.
Renée Mie Fredensborg Hansen, Eero Rinne, Sinéad Louise Farrell, and Henriette Skourup
The Cryosphere, 15, 2511–2529, https://doi.org/10.5194/tc-15-2511-2021, https://doi.org/10.5194/tc-15-2511-2021, 2021
Short summary
Short summary
Ice navigators rely on timely information about ice conditions to ensure safe passage through ice-covered waters, and one parameter, the degree of ice ridging (DIR), is particularly useful. We have investigated the possibility of estimating DIR from the geolocated photons of ICESat-2 (IS2) in the Bay of Bothnia, show that IS2 retrievals from different DIR areas differ significantly, and present some of the first steps in creating sea ice applications beyond e.g. thickness retrieval.
Luisa von Albedyll, Christian Haas, and Wolfgang Dierking
The Cryosphere, 15, 2167–2186, https://doi.org/10.5194/tc-15-2167-2021, https://doi.org/10.5194/tc-15-2167-2021, 2021
Short summary
Short summary
Convergent sea ice motion produces a thick ice cover through ridging. We studied sea ice deformation derived from high-resolution satellite imagery and related it to the corresponding thickness change. We found that deformation explains the observed dynamic thickness change. We show that deformation can be used to model realistic ice thickness distributions. Our results revealed new relationships between thickness redistribution and deformation that could improve sea ice models.
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.
Stephan Paul and Marcus Huntemann
The Cryosphere, 15, 1551–1565, https://doi.org/10.5194/tc-15-1551-2021, https://doi.org/10.5194/tc-15-1551-2021, 2021
Short summary
Short summary
Cloud cover in the polar regions is difficult to identify at night when using only thermal-infrared data. This is due to occurrences of warm clouds over cold sea ice and cold clouds over warm sea ice. Especially the standard MODIS cloud mask frequently tends towards classifying open water and/or thin ice as cloud cover. Using a neural network, we present an improved discrimination between sea-ice, open-water and/or thin-ice, and cloud pixels in nighttime MODIS thermal-infrared satellite data.
Stephen E. L. Howell, Randall K. Scharien, Jack Landy, and Mike Brady
The Cryosphere, 14, 4675–4686, https://doi.org/10.5194/tc-14-4675-2020, https://doi.org/10.5194/tc-14-4675-2020, 2020
Short summary
Short summary
Melt ponds form on the surface of Arctic sea ice during spring and have been shown to exert a strong influence on summer sea ice area. Here, we use RADARSAT-2 satellite imagery to estimate the predicted peak spring melt pond fraction in the Canadian Arctic Archipelago from 2009–2018. Our results show that RADARSAT-2 estimates of peak melt pond fraction can be used to provide predictive information about summer sea ice area within certain regions of the Canadian Arctic Archipelago.
Hoyeon Shi, Byung-Ju Sohn, Gorm Dybkjær, Rasmus Tage Tonboe, and Sang-Moo Lee
The Cryosphere, 14, 3761–3783, https://doi.org/10.5194/tc-14-3761-2020, https://doi.org/10.5194/tc-14-3761-2020, 2020
Short summary
Short summary
To estimate sea ice thickness from satellite freeboard measurements, snow depth information has been required; however, the snow depth estimate has been considered largely uncertain. We propose a new method to estimate sea ice thickness and snow depth simultaneously from freeboards by imposing a thermodynamic constraint. Obtained ice thicknesses and snow depths were consistent with airborne measurements, suggesting that uncertainty of ice thickness caused by uncertain snow depth can be reduced.
Nicholas C. Wright, Chris M. Polashenski, Scott T. McMichael, and Ross A. Beyer
The Cryosphere, 14, 3523–3536, https://doi.org/10.5194/tc-14-3523-2020, https://doi.org/10.5194/tc-14-3523-2020, 2020
Short summary
Short summary
This work presents a new dataset of sea ice surface fractions along NASA Operation IceBridge flight tracks created by processing hundreds of thousands of aerial images. These results are then analyzed to investigate the behavior of meltwater on first-year ice in comparison to multiyear ice. We find preliminary evidence that first-year ice frequently has a lower melt pond fraction than adjacent multiyear ice, contrary to established knowledge in the sea ice community.
Wolfgang Dierking, Harry L. Stern, and Jennifer K. Hutchings
The Cryosphere, 14, 2999–3016, https://doi.org/10.5194/tc-14-2999-2020, https://doi.org/10.5194/tc-14-2999-2020, 2020
Short summary
Short summary
Monitoring deformation of sea ice is useful for studying effects of ice compression and divergent motion on the ice mass balance and ocean–ice–atmosphere interactions. In calculations of deformation parameters not only the measurement uncertainty of drift vectors has to be considered. The size of the area and the time interval used in the calculations have to be chosen within certain limits to make sure that the uncertainties of deformation parameters are smaller than their real magnitudes.
Igor E. Kozlov, Evgeny V. Plotnikov, and Georgy E. Manucharyan
The Cryosphere, 14, 2941–2947, https://doi.org/10.5194/tc-14-2941-2020, https://doi.org/10.5194/tc-14-2941-2020, 2020
Short summary
Short summary
Here we demonstrate a recently emerged opportunity to retrieve high-resolution surface current velocities from sequential spaceborne radar images taken over low-concentration ice regions of polar oceans. Such regularly available data uniquely resolve complex surface ocean dynamics even at small scales and can be used in operational applications to assess and predict transport and distribution of biogeochemical substances and pollutants in ice-covered waters.
Jeong-Won Park, Anton Andreevich Korosov, Mohamed Babiker, Joong-Sun Won, Morten Wergeland Hansen, and Hyun-Cheol Kim
The Cryosphere, 14, 2629–2645, https://doi.org/10.5194/tc-14-2629-2020, https://doi.org/10.5194/tc-14-2629-2020, 2020
Short summary
Short summary
A new Sentinel-1 radar-based sea ice classification algorithm is proposed. We show that the readily available ice charts from operational ice services can reduce the amount of manual work in preparation of large amounts of training/testing data and feed highly reliable data to the trainer in an efficient way. Test results showed that the classifier is capable of retrieving three generalized cover types with overall accuracy of 87 % and 67 % in the winter and summer seasons, respectively.
Marcel König and Natascha Oppelt
The Cryosphere, 14, 2567–2579, https://doi.org/10.5194/tc-14-2567-2020, https://doi.org/10.5194/tc-14-2567-2020, 2020
Short summary
Short summary
We used data that we collected on RV Polarstern cruise PS106 in summer 2017 to develop a model for the derivation of melt pond depth on Arctic sea ice from reflectance measurements. We simulated reflectances of melt ponds of varying color and water depth and used the sun zenith angle and the slope of the log-scaled reflectance at 710 nm to derive pond depth. We validated the model on the in situ melt pond data and found it to derive pond depth very accurately.
Cited articles
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
Buus-Hinkler, J., Wulf, T., Stokholm, A. R., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Solberg, R., Longépé, N., and Brandt Kreiner, M.: AI4Arctic Sea Ice Challenge Dataset, DTU [code and data set], https://doi.org/10.11583/DTU.c.6244065.v2, 2022. a, b, c
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, 2020. a
Chen, X., Scott, K. A., Jiang, M., Fang, Y., Xu, L., and Clausi, D. A.: Sea Ice Classification With Dual-Polarized SAR Imagery: A Hierarchical Pipeline, in: Proc. IEEE/CVF Winter Conf. Appl. Comput. Vis., Waikoloa, USA, January, 2023, 224–232, 2023a. a
Chen, X., Valencia, R., Soleymani, A., and Scott, K. A.: Predicting Sea Ice Concentration With Uncertainty Quantification Using Passive Microwave and Reanalysis Data: A Case Study in Baffin Bay, IEEE Trans. Geosci. Remote Sens., 61, 1–13, https://doi.org/10.1109/TGRS.2023.3250164, 2023b. a, b
Chi, J., Kim, H.-c., Lee, S., and Crawford, M. M.: Deep learning based retrieval algorithm for Arctic sea ice concentration from AMSR2 passive microwave and MODIS optical data, Remote Sens. Environ., 231, 111204, https://doi.org/10.1016/j.rse.2019.05.023, 2019. a
Cooke, C. L. and Scott, K. A.: Estimating sea ice concentration from SAR: Training convolutional neural networks with passive microwave data, IEEE Trans. Geosci. Remote, 57, 4735–4747, 2019. a
De Gelis, I., Colin, A., and Longépé, N.: Prediction of categorized Sea Ice Concentration from Sentinel-1 SAR images based on a Fully Convolutional Network, IEEE J. Sel. Top. Appl., 14, 5831–5841, https://doi.org/10.1109/JSTARS.2021.3074068, 2021. 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
Huang, Y., Ren, Y., and Li, X.: Classifying Sea Ice Types from SAR Images Using a U-Net-Based Deep Learning Model, in: 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, Brussels, Belgium, July 2021, 3502–3505, https://doi.org/10.1109/IGARSS47720.2021.9554511, 2021. 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., 15, 7964–7977, https://doi.org/10.1109/JSTARS.2022.3205849, 2022. 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, 2021a. a
Khaleghian, S., Ullah, H., Krmer, T., Eltoft, T., and Marinoni, A.: Deep Semi-Supervised Teacher-Student Model based on Label Propagation for Sea Ice Classification, IEEE J. Sel. Top. Appl., 14, 10761–10772, https://doi.org/10.1109/JSTARS.2021.3119485, 2021b. 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, https://doi.org/10.1109/TGRS.2022.3144731, 2022. a
Kruk, R., Fuller, M. C., Komarov, A. S., Isleifson, D., and Jeffrey, I.: Proof of concept for sea ice stage of development classification using deep learning, Remote Sens., 12, 2486, https://doi.org/10.3390/rs12152486, 2020. 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
Li, X.-M., Qiu, Y., Wang, Y., Huang, B., Lu, H., Chu, M., Fu, H., and Hui, F.: Light from space illuminating the polar silk road, Int. J. Digit. Earth, 15, 2028–2045, 2022. a
Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Dollár, P.: Focal loss for dense object detection, in: Proceedings of the IEEE international conference on computer vision, Venice, Italy, 2017, 2980–2988, https://doi.org/10.1109/ICCV.2017.324, 2017. a
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., 14, 10919–10928, https://doi.org/10.1109/JSTARS.2021.3122546, 2021a. a, b
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows, in: Proceedings of the IEEE/CVF international conference on computer vision, Montreal, QC, Canada, 2021, 10012–10022, https://doi.org/10.1109/ICCV48922.2021.00986, 2021b. a
Loshchilov, I. and Hutter, F.: SGDR: Stochastic gradient descent with warm restarts, arXiv [preprint], https://doi.org/10.48550/arXiv.1608.03983, 2016. 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, 2022a. a
Lyu, H., Huang, W., and Mahdianpari, M.: A Meta-Analysis of Sea Ice Monitoring Using Spaceborne Polarimetric SAR: Advances in the Last Decade, IEEE J. Sel. Top. Appl., 15, 6158–6179, 2022b. a
Mahmud, M. S., Nandan, V., Singha, S., Howell, S. E., Geldsetzer, T., Yackel, J., and Montpetit, B.: C-and L-band SAR signatures of Arctic sea ice during freeze-up, Remote Sens. Environ., 279, 113129, https://doi.org/10.1016/j.rse.2022.113129, 2022. a
Minnett, P., Alvera-Azcárate, A., Chin, T., Corlett, G., Gentemann, C., Karagali, I., Li, X., Marsouin, A., Marullo, S., Maturi, E., Santoleri, R., Saux Picart, S., Steele, M., and Vazquez-Cuervo, J.: Half a century of satellite remote sensing of sea-surface temperature, Remote Sens. Environ., 233, 111366, https://doi.org/10.1016/j.rse.2019.111366, 2019. 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, b
Ren, Y., Li, X., Yang, X., and Xu, H.: Development of a Dual-Attention U-Net Model for Sea Ice and Open Water Classification on SAR Images, IEEE Geosci. Remote Sens. Lett., 19, 1–5, https://doi.org/10.1109/LGRS.2021.3058049, 2021. a
Ronneberger, O., Fischer, P., and Brox, T.: U-Net: Convolutional networks for biomedical image segmentation, in: Intl. Conf. Med. Image Comput. Comput. Assist. Interv., Munich, Germany, 5–9 October 2015, 234–241, https://doi.org/10.1007/978-3-319-24574-4_28, 2015. a
Saldo, R., Kreiner, M. B., Buus-Hinkler, J., Pedersen, L. T., Malmgren-Hansen, D., Nielsen, A. A., and Skriver, H.: AI4Arctic/ASIP Sea Ice Dataset – version 2, DTU [data set], https://doi.org/10.11583/DTU.13011134.v3, 2021. a
Shamshiri, R., Eide, E., and Høyland, K. V.: Spatio-temporal distribution of sea-ice thickness using a machine learning approach with Google Earth Engine and Sentinel-1 GRD data, Remote Sens. Environ., 270, 112851, https://doi.org/10.1016/j.rse.2021.112851, 2022. a
Soleymani, A. and Scott, K. A.: Evaluation of a Neural Network on Sea Ice Concentration Estimation in MIZ Using Passive Microwave Data, in: IGARSS, Brussels, Belgium, 2021, 5656–5659, https://doi.org/10.1109/IGARSS47720.2021.9553638, 2021. a
Song, W., Li, M., Gao, W., Huang, D., Ma, Z., Liotta, A., and Perra, C.: Automatic Sea-Ice Classification of SAR Images Based on Spatial and Temporal Features Learning, IEEE Trans. Geosci. Remote, 59, 9887–9901, https://doi.org/10.1109/TGRS.2020.3049031, 2021. a
Stokholm, A., Wulf, T., Kucik, A., Saldo, R., Buus-Hinkler, J., and Hvidegaard, S. M.: AI4SeaIce: Toward Solving Ambiguous SAR Textures in Convolutional Neural Networks for Automatic Sea Ice Concentration Charting, IEEE Trans. Geosci. Remote, 60, 1–13, https://doi.org/10.1109/TGRS.2022.3149323, 2022. a, b, c
Stokholm, A., Buus-Hinkler, J., Wulf, T., Korosov, A., Saldo, R., Arthurs, D., Solberg, R., Longépé, N., and Kreiner, M.: The AutoICE Competition: Automatically Mapping Sea Ice in the Arctic, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-13038, https://doi.org/10.5194/egusphere-egu23-13038, 2023. a. a
Stokholm, A., Kucik, A., Longépé, N., and Hvidegaard, S. M.: AI4SeaIce: Task Separation and Multistage Inference CNNs for Automatic Sea Ice Concentration Charting, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-976, 2023b. a
Stokholm, A. R., Buus-Hinkler, J., Wulf, T., Korosov, A., Saldo, R., Pedersen, L. T., Arthurs, D., Dragan, I., Modica, I., Pedro, J., Debien, A., Chen, X., Patel, M., Cantu, F. J. P., Turnes, J. N., Park, J., Xu, L., Scott, A. K., Clausi, D. A., Fang, Y., Jiang, M., Taleghanidoozdoozan, S., Brubacher, N. C., Soleymani, A., Gousseau, Z., Smaczny, M., Kowalski, P., Komorowski, J., Rijlaarsdam, D., van Rijn, J. N., Jakobsen, J., Rogers, M. S. J., Hughes, N., Zagon, T., Solberg, R., Longépé, N., and Kreiner, M. B.: The AutoICE Challenge, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-2648, 2023c. 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 Trans. Geosci. Remote, 54, 4524–4533, https://doi.org/10.1109/TGRS.2016.2543660, 2016. a
Wang, L., Scott, K. A., and Clausi, D. A.: 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, 2017. a
Zhang, T., Yang, Y., Shokr, M., Mi, C., Li, X.-M., Cheng, X., and Hui, F.: Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data, Remote Sensing, 13, 1452, https://doi.org/10.3390/rs13081452, 2021. a
Zhang, Y., Zhu, T., Spreen, G., Melsheimer, C., Huntemann, M., Hughes, N., Zhang, S., and Li, F.: Sea ice and water classification on dual-polarized Sentinel-1 imagery during melting season, The Cryosphere Discuss. [preprint], https://doi.org/10.5194/tc-2021-85, 2021. a
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.
This paper introduces an automated sea ice mapping pipeline utilizing a multi-task U-Net...
Special issue