Oceanviewer Banner

Improved Sea Ice Prediction Through Assimilation of Ice Thickness Info and SAR Image Classification

  • Headshot

    Andrea Scott University of Waterloo

Developing automated methods to assimilate data from synthetic aperture radar (SAR) sensors, which provide higher resolution information about the sea ice state.

The focus of the project is to studying changes in sea ice thickness and extreme ice features and to develop tools to help stakeholders forecast these conditions.

Sea ice thickness data acquired through an airborne survey in the Beaufort Sea last spring, in addition to joint surveys with collaborators from the University of Alaska, supported the emerging picture indicating ice is becoming more mobile and more hazardous. This result has generated international interest. The airborne data was acquired with coincident synthetic aperture radar (SAR) satellite imagery, the main type of data used to generate maps of ice type (related to ice thickness) provided by operational services.

Work is ongoing to develop automated methods to retrieve sea ice information directly from SAR images. In this context, a convolutional neural network was developed that is able to retrieve ice concentration accurately. To make use of SAR retrievals in sea ice forecasts, methods are being developed to assimilate ice concentration, thickness, and ice type, with emphasis on improving the representation of sharp features (e.g. ridges), which may help predict pressurized ice, a major factor in vessel besetting.


  • Canatec Associates International Ltd.
  • Environment Canada
  • Environment Canada - Canadian Ice Service
  • Exxon Mobil
  • Imperial Oil


  • Matt Arkett
  • Mark Buehner
  • Tom Carrieres
  • David Clausi
  • Neil Darlow
  • Claudde Duguay
  • Hajo Eicken
  • Christian Haas
  • Kumaraswamy Ponnambalam University of Waterloo


  • Nazanin Asadi University of Waterloo
  • Alec Casey York University
  • Kevin Kang University of Waterloo
  • Homa Kheyrollah Pour University of Waterloo
  • Lei Wang University of Waterloo Ontario


  • Clausi,David,Li,Fan,Wang,Lei ,Xu, L.. 2015, A semi-supervised approach for ice-water classification using dual-polarization SAR satellite imagery, Computer Vision and Pattern Recognition (CVPR) Workshop on EARTHVISION,
  • Scott, A., Liu,J., Fieguth, P.. 2015, Curvelet based feature extraction of dynamic ice from SAR imagery, IEEE International Geoscience and Remote Sensing Symposium,
  • Mussells, O. 2015, Observing pressured sea ice in the Hudson Strait using RADARSAT: Implications for shipping, MSc Thesis, Department of Geography, University of Ottawa, 122 pp.,
  • Azetsu-Scott,Kumiko,Mozaffari A., Azad, N.L. and Chenouri, S.. 0, A hierarchical selective ensemble randomized neural network hybridized with heuristic feature select, Applied Intelligence, 46: 16. ,10.1007/s10489-016-0815-x.
  • Azetsu-Scott,Kumiko,Mozaffari, A., Azad, N.L. and Chenouri, S. A modular. 0, A modular ridge extreme learning machine with differential evolutionary distributor applied to the estimation of sea ice thickness, Soft Computing, 10.1007/s00500-016-2074-5.
  • Azetsu-Scott,Kumiko,Ponnambalam,Kumaraswamy,Sun F.. 0, Arctic ice thickness prediction for ship operation under hierarchical Bayesian network data assimilation framework, International Conference on Inverse Problems in Engineering,
  • Buehner,Mark,Carrieres,Tom,Scott,Andrea, Ashouri, Z., Pogson, L.,. 2015, Assimilation of Ice and Water Observations from SAR Imagery to Improve Estimates of Sea Ice Concentration , Tellus A,
  • Li,Fan,. 2015, Automated Remote Sensing Image Interpretation with Limited Labeled Training Data, Ph.D. Thesis, Systems Design Engineering, University of Waterloo, 102 pp.,
  • Azetsu-Scott,Kumiko,Liu, J., Scott, and Fieguth, P.. 0, Automatic Detection of the Ice Edge in SAR imagery using Curvelet, Remote Sensing, 8, 480,10.3390/rs8060480..
  • Christian,Jim,N Steiner, W Lee. 2013, Enhanced Gas Fluxes in Small Sea Ice Leads and Cracks 0 Effects on CO2 Exchange and Ocean Acidification, Journal of Geophysical Research,
  • Christian,Jim,N Steiner, K Six et al. . 2013, Future Ocean Acidification in the Canada Basin and Surrounding Arctic Ocean from CMIP5 Earth System Models, Journal of Geophysical Research,
  • Bublitz,Anne,Casey,Alec,Haas,Christian, J. Beckers. 2016, Ice Thickness in the Beaufort Sea and Northwest Passage in April 2016 and comparison with April 2015, Sea Ice Outlook preseason contribution,
  • Clausi,David,Scott,Andrea,Wang,Lei ,. 2016, Ice concentration estimation during melting period of the Arctic from dual-polarized SAR images using deep convolutional neural networks, IEEE Transactions on Geoscience and Remote Sensing, TGRS-2015-00060,
  • Azetsu-Scott,Kumiko,Clausi,David,Wang,Lei ,. 0, Ice concentration estimation in the Gulf of St. Lawrence using a fully convolutional neural network, IEEE,
  • Azetsu-Scott,Kumiko,Wang,Lei ,Clausi, D.. 0, Ice concentration estimation in the Gulf of St. Lawrence using a fully convolutional neural network, IEEE Geoscience and Remote Sensing Symposium,
  • Haas,Christian,Howell,Stephen,. 2015, Ice thickness in the Northwest Passage, Geophyical Research Letters, 42, 10.1002/2015GL065704.
  • Clausi,David,Scott,Andrea,Wang,Lei ,. 2016, Improved Sea Ice Concentration Estimation Through Fusing Classified SAR Imagery and AMSR-E Data, Canadian Journal of Remote Senisng, 10:1080/07038992.2016.1152547.
  • Dawson,Jackie,Mussells, O., and Howell, S.E.. 2017, Navigating Pressured Ice: risks and hazards for winter resource-based shipping in the Canadian Arctic, Oceans and Coastal Management,
  • Haas,Christian,Beckers, J., J.A. Casey. 2017, Retrievals of lake ice thickness from Great Slave and Great Bear Lakes using CryoSat-2, IEEE TGARS,
  • Haas,Christian,Kaleschke, L., X. Tian-Kunze, N. Maaß, A. Beitsch, A. Wernecke, M. Miernecki, G. Müller, B.H. Fock, A.MU Gierisch, K.H. Schlünzen, T. Pohlmann, M. Dobrynin, S. Hendricks, J. Asseng, R. Gerdes, P. Jochmann, N. Reimer, J. Holfort, C. Melsheimer, G. Heygster. 2016, SMOS sea ice product: Operational application and validation in the Barents Sea marginal ice zone, Remote Sensing of Environment, 10.1016/j.rse.2016.03.009.
  • Haas,Christian,Beckers, J.F., A.H.H. Renner, G. Spreen, S. Gerland. 2015, Sea ice surface roughness estimates from airborne laser scanner and laser altimeter observations in Fram Strait and north of Svalbard, Annals of Glaciology, 56(69), 10.3189/2015AoG69A717.
  • Casey,Alec,Haas,Christian,Howell,Stephen,A. Tivy. 2015, Separability of sea ice types from wide swath C- and L-band synthetic aperture radar imagery acquired during summer melt, Remote Sensing of Environment, 174, 314–328,10.1016/j.rse.2015.12.021.
  • Copeland,Luke,Dawson,Jackie,Pizzolato,Larissa,S. E. L. Howell, F. Laliberté. 2016, The influence of declining sea ice on shipping activity in the Canadian Arctic, Geophys. Res. Lett, 43, 12,146–12,154,10.1002/2016GL071489.
  • Dawson,Jackie,Mussells, O., and Howell, S.E.L.. 2016, Using RADARSAT to identify sea ice ridges and their implications for shipping in the Hudson Strait, Arctic, 69(4),10.14430/arctic4604 .

Accurate information about sea ice conditions is critical for weather forecasting and for safe operations of industries such as fishing, shipping, and oil exploration in Canadian waters.