New image analysis technologies for fast and accurate retrieval of sea ice floe size distribution (FSD) from satellite SAR imagery

Lead Research Organisation: Scottish Association For Marine Science
Department Name: Scottish Association For Marine Science


Arctic sea ice is changing rapidly. The most profound example is during the summer of 2012, in which the lowest ice extent was recorded since satellite sensors began to monitor sea ice in 1979. Within just one week, as a violent storm passed the Arctic in August of 2012, sea ice area equivalent to nearly twice the size of UK (0.4 million square kilometres) disappeared, leaving ice-free water up to 80 N by mid-August in western Arctic.

One of the key processes that cause such rapid sea ice decline is sea-ice floe breakup during the winter-to-summer transition. During this transition the edge of sea ice retreats to the north, exposing larger open water fetch to generate waves, propagating into the ice pack, allowing larger sea-ice floes to break into smaller ones. As the floes become smaller, they melt faster and become more dynamic. At the same time more solar radiation is absorbed through exposed open water areas, which makes the upper ocean layer warmer and in turn promotes faster ice melting. This chain reaction can accelerate the sea ice retreat and thus impact the minimum ice extent.

This important floe breakup and associated effects are poorly implemented in sea-ice/climate models. This is partly due to lack of understanding and verification of our current knowledge on the processes as well as due to complexity of the processes that makes it difficult to effectively implement them into "simple" representations in the models. Producing effective parameterisations requires accurate data on in-situ floe size distribution (FSD) that can be used to verify and refine the known parameterisations as well as to formulate new ones.

Satellite Synthetic Aperture Radar (SAR) provides observations of sea ice unhindered by either darkness or cloud, thus provide ideal raw data from which FSD can be retrieved from dark winter to cloudy summer in the Arctic. There is an increasing number of satellite SAR images being acquired in the Arctic, and often at spatial resolutions in the images as good as 1-20 m. More importantly satellite SAR images are being acquired over autonomous buoy systems and in conjunction with field campaigns. This provides the ideal framework to measure the full range of ocean, sea-ice and atmosphere parameters to investigate complex floe breakup process.

However the challenge we have is a lack of proven-quality algorithms that can derive FSD from satellite SAR images fast and accurately. Thresholding algorithms previously applied to the problem are not adequate for quantitative analysis and the performance has not been precisely assessed.

In this project we, for the first time, combine sea ice physics with edge-cutting image processing techniques to develop FSD algorithms at a completely different level. We leverage the latest image processing technologies which include a) wavelet algorithms to reduce the speckle noise while increasing the contrast of the boundary between ice and water, b) local-statistics based algorithm to extract ice floe features from the background open water, c) and a combination of edge-preserving watershed and split-and-merge algorithms to effectively split up the touching boundary of the floes. We expect this set of new algorithms will produce much more accurate FSD from satellite SAR images, and lay a foundation develop universal algorithm that can be used to build a long-term sea-ice FSD database.

Planned Impact

1. Sea ice floe breakup and its cascading effects are still poorly parameterised in the sea-ice/climate models. This is partly because of complexity of the process that makes it challenging to represent them as "simple" parameterisations in the models as well as partly because of lack of understanding and verification of our current knowledge on this process.

2.The main driver of this project is a lack of proven-quality algorithm(s) that can derive FSD fast and accurately from satellite SAR. By developing such algorithm(s) through this TPoC project, it will benefit wide range of sea-ice and climate research groups as well as private sectors for better ice management.

3.Through this project we will build a cohesive user/research group that has common interested in FSD retrieval algorithm, the associated science and commercial interests. An impact workshop will be carefully organized to achieve this cohesive group. It is important that the group comprise diverse backgrounds of sea-ice physics, modeling, image processing, commercial partners. This multi-faceted but focused group can be the key output of our impact plan.

Full details of the impacts of this project are outlines in Pathway to Impact document.


10 25 50
Description We developed a unique image processing algorithm that can automatically analysis satellite synthetic aperture radar (SAR) images to obtain sea ice information. This allowed us to estimate the temporal and spatial variability of how sea ice change its size in the Arctic Ocean. This study revealed two important insights: (i) winter ice types and deformation (resulting leads/cracks) affect the summer floe size evolution; (ii) while floe breakup events are typically associated with strong wind events, floe breakup can occur due to severe (vertical) melting of thinner ice types serving as weak fracture points even with relatively moderate wind forcing. These results indicate how important it is to measure and quantify how winter ice types/features and deformations transform into spring ice floes, and how these spring ice floes evolve through summer melt and breakup.
Exploitation Route Journal papers describing the developed algorithms in detail have been published, so that any relevant scientists/engineers can access to the information. Currently we are planning to conduct an algorithm sensitivity study, led by Dr Horvat (to have objective assessment). This sensitivity study will provide more detailed evaluation of the performance of the algorithms developed through this project. Our findings and technology have been used to support climate/sea ice-ocean modelling scientists by providing new data sets of sea ice floe size distribution in the Arctic.
Sectors Aerospace, Defence and Marine,Education,Environment

Description NERC Standard grants
Amount £311,966 (GBP)
Funding ID NE/R000654/1 
Organisation Natural Environment Research Council 
Sector Public
Country United Kingdom
Start 10/2017 
End 03/2020
Title FSD algorithm 
Description The funding enabled us to develop a unique algorithm to process satellite Synthetic Aperture Radar (SAR) image data to derive sea ice floe size distribution. 
Type Of Material Computer model/algorithm 
Provided To Others? No  
Impact The algorithm developed has been used to derive sea ice floe size distribution from satellite SAR imagery, and the derived data have been used to calibrate and validate numerical sea ice models (Zhang et al. 2016). 
Title FSD data to UK Polar Data Centre 
Description This data set generated from the algorithm developed through two NERC grants received. The data set contains sea-ice floe size distribution data derived from USGS GFL high-resolution imagery (see below for the title of data set and the data centre stored). The data set is under embargo until March 2021. Hwang, B. (2020). Sea-ice floe size distribution data derived from USGS GFL high-resolution satellite imagery for the pre-ponding period of 2000-2014 [Data set]. UK Polar Data Centre, Natural Environment Research Council, UK Research & Innovation. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? No  
Impact This data set is unique, containing sea-ice floe size distribution over 10 years at three different locations across the Arctic. 
Title Implementation of image processing algorithm to forest research in Africa 
Description The computer algorithm developed for the analysis of Arctic sea ice floe size has been implemented to analyse the canopy cover of trees in Ethiopian forests. 
Type Of Material Computer model/algorithm 
Year Produced 2019 
Provided To Others? No  
Impact The algorithm allowed to process upward-looking camera photos to analyse canopy cover in Ethiopian coffee forest. This development now allows us to assess biodiversity and coffee forest mapping: the impact beyond the originally intended for the Arctic research. 
Description Collaboration with ETS 
Organisation Superior Technology School
Country Canada 
Sector Academic/University 
PI Contribution Through the NERC project (NE/M00600X/1) I was able to visit Dr. Ben Ayed to discuss about future project to apply advanced optimization algorithms to sea ice feature extraction problems.
Collaborator Contribution During the visit Dr. Ben Ayed contributed to the discussion regarding the latest development of optimization algorithms which can be applied to specifically sea ice feature extraction. He contributed to a co-authored paper published in 2017.
Impact As a results of discussion, the collaborative proposal was formulated and successfully submitted to US Office of Naval Research (the funding decision is still pending). The collaboration resulted in a co-authored publication.
Start Year 2014
Description Collaboration with U of Strathclyde 
Organisation University of Strathclyde
Country United Kingdom 
Sector Academic/University 
PI Contribution The NERC project contributed Drs Ren, Marshall and Murray to join the inter-disciplinary workshop (called FSD workshop, directly funded by NE/M00600X/1).
Collaborator Contribution Dr. Ren (Co-I of the NERC project) contributed to the coordination of the workshop. Dr Ren contributed to two co-authored papers published in 2017. Drs Marshall and Murray contributed to the collaborative workshop by providing the presentation on their image processing works and participating group discussions regarding future collaborative research involving image processing and observations of sea ice.
Impact The collaborative workshop (called FSD workshop, held on July 6-7, 2015 at the Scottish Association for Marine Science). The collaboration resulted in two co-authored papers in 2017, and one NERC proposal (Dr Ren as CoI) in 2018.
Start Year 2015
Description Guest speaker at SOLAS Workshop on Remote Sensing for Studying the Ocean-Atmosphere Interface 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact I was invited as a guest speaker at the SOLAS (Surface Ocean Lower Atmosphere Study) Workshop on Remote Sensing for Studying the Ocean-Atmosphere Interface, USA. In this talk, I introduced my research on Marginal Ice Zone (MIZ) from the algorithm development to process studies. This talk raised the importance of MIZ process related research and discussions and led to future related activity within SOLAS community: forming a discussion session on MIZ process at the SOLAS Open Science Conference to be held in Japan in April 2019.
Year(s) Of Engagement Activity 2012,2018
Description Invited talk at the Beijing University of Aeronautics and Astronautics, Beijing, China 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Postgraduate students
Results and Impact Invited talk at the Beijing University of Aeronautics and Astronautics. We shared our findings with Chinese professors and postgradudate students, and discussed about future collobrations.
Year(s) Of Engagement Activity 2017
Description Talk at the Isaac Newton Institute for Mathematical Sciences, University of Cambridge, UK 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact The workshop was organised by international leading scientists in sea ice modelling and observation. More than 50 scientists were attended at the workshop. This provided an opportunity to introduce our findings and enhance further collaborations.
Year(s) Of Engagement Activity 2017