Denis Demchev (OSIRS group) and Valeria Selyuzhenok (Zubrov State Oceanographic Institute, Russia) recently published a study introducing a fully-automated method to track fast ice and stamukhi in Arctic coastal zones. Their method is faster and more objective than manually-produced operational sea-ice charts and can provide near-real time information!

Fast ice and stamukhi in the Arctic

The Arctic coastline is strongly affected by seasons. In the summer it is ice-free and over the course of the autumn and winter months, something called fast ice forms along it. Here, “fast” does not mean “quick”, but rather “attached and immobile”.  Ice is fastened to the coastline at temperatures well below freezing, and it cannot be moved by winds or waves. It is frozen stiff in place, attached to the land. How wide this zone spans from the land towards the open ocean depends on local factors such as ocean depth, regional winds, and currents.  

At the other end of it, towards the open ocean, it looks different. There, it goes either from one piece of ice to many ice floes floating loosely in the water. Or stamukhi are formed: A stamukha is a pile of many broken sea ice bits, which does not just reach upwards, but also downwards into the water, even down into the seafloor itself. This pile usually extends as a belt parallel to the coast. Both fast ice and stamukhi can be detected from space, satellite imagery is invaluable here. But why is it necessary to know about the ice extent and locations of stamukhi?

Benefits and applications of Arctic coastline surveillance

Fast ice and stamukhi pose a risk to coastal infrastructure. Offshore pipelines and cables leading to the land can be damaged by these downward-reaching ice rubble piles, the stamukhi. Surveilling them from space by producing sea-ice maps helps mitigate that risk.

Another application for sea-ice maps is traffic: Ships are not indestructible, so choosing the safest route for ship traffic in the Arctic is crucial.

Aside from the immediate safety benefits, surveilling the Arctic coastline benefits climatology and ecology. Warming in the Arctic impacts how long fast ice is present along Arctic coastlines – the ice-free period increases. During the summer, wind and waves can affect the shore more drastically, more erosion occurs along it, leading to land loss along the coast.

This surveillance happens today by specialists producing sea-ice charts manually.

Manual sea-ice charts

A comparison of manually produced sea-ice maps (in light blue, by the Arctic and Antarctic Research Institute, Russia) and the automatically produced sea-ice edge outputs (blue lines, by the new method), for four different periods in the winter of 2007. Figure 5 in Selyuzhenok and Demchev (2021)

These sea-ice chart specialists use satellite images and in-situ data directly from the Arctic as basis for their work. In Russia for example, the Arctic and Antarctic Research Institute is responsible for this work. They release weekly forecasts, showing the sea ice for 2-5 days into the future. The production is very labour-intensive, and it can be subjective since humans are at work here. Aside from this common method, some semi-automatic approaches do exist, making use of optical satellite images. So, why do we need another way of producing sea-ice charts to Arctic coastline surveillance?

Fully-automated sea-ice tracking

Selyuzhenok and Demchev developed a method to circumvent these drawbacks of manual or semi-automatic approaches. Conventional optical satellite imagery cannot depict the Earth’s surface when it is dark, or when clouds cover the surface. This new method by Selyuzhenok and Demchev utilizes synthetic aperture radar (SAR) data, and this kind of satellite-based “imagery” can see both through clouds and darkness. The Earth’s surface can be observed with it during night and day, with almost any weather conditions present and in very high resolution! The authors combined SAR data with mathematical formulas to describe how sea ice drifts in a natural way, and their approach was successful. They compared their automatically-created output with operational sea-ice charts for the same time frame, and they show that their approach is capable of accurately detecting the fast ice edge and the location of stamukhi along Arctic coastlines (see the figure on the side)! This is exciting for any scientific field where Arctic coastline surveillance has applications.

Co-author Denis Demchev on the importance of this study: “Accurate fast ice delineation is of big importance for scientific analysis, ice charting and other practical needs. Fast ice is a great challenge in sea-ice modelling that makes the observational data crucial also in this field. The proposed method is an attempt aimed to make fast ice mapping objective, robust and time efficient by combining benefits of SAR data and state-of-the-art sea-ice tracking algorithms.”

Reference:

Selyuzhenok, V, & D Demchev. An Application of Sea Ice Tracking Algorithm for Fast Ice and Stamukhas Detection in the Arctic. Remote Sensing 2021. 13(18), 3783, https://doi.org/10.3390/rs13183783