Editors’ Vox is a blog from AGU’s Publications Department.

Antarctic sea ice extent has plummeted throughout 2023, with its rapid loss surprising scientists across the world. Meanwhile, attention is being drawn to the importance of landfast sea ice (fast ice), which is “locked in” against and around the Antarctic coast, like a seatbelt around the continent.

Fast ice is experiencing a decline of its own, as highlighted in a recent study in Reviews of Geophysics, which is bad news for the wide variety of Earth system processes linked to it. We asked some of the authors to give an overview of landfast ice, how scientists study it, and what is still unknown.

In simple terms, what is landfast ice and where does it occur?

Landfast ice, also confusingly abbreviated to “fast ice” (despite it not moving!), is sea ice which has formed from frozen seawater, but does not move in response to the currents and winds, unlike pack ice. Fast ice can be fastened either to the coastline itself, or, commonly, icebergs which run aground on the relatively shallow Antarctic continental shelf, at depths of up to about 400 meters. Generally, it forms a discontinuous belt around the Antarctic coastline, with a width of 50 to 200 kilometers. Its area generally varies between about 221,000 and 601,000 km2, so it forms a significant fraction of overall sea ice (about 3 to 19 million km2).

Why is it important to understand fast ice as distinct from other forms of sea ice? 

Fast ice has a variety of distinctly different roles in the physical, biogeochemical, and ecological systems when compared to pack ice. For example, due to its immobility, fast ice:

  • Pushes mobile pack ice further to the north, thus contributing to the formation of polynyas. Polynyas, or regions of lower-than-expected ice concentration, are important for the global climate since they are regions of intensive heat transfer from the “warm” ocean (–1.8 ºC) to the cold atmosphere, because there is no sea ice present to act as an insulator. Thus, polynyas are regions of very intensive sea ice production. As new ice forms, salt gets rejected back into the water, because ice is fresher than the ocean. This water is very dense, and can sink to the bottom of the ocean, thereby driving global ocean circulation. 
  • Is generally thicker than pack ice, and acts as a reservoir of nutrients, e.g., iron, which is crucial for phytoplankton growth in the Southern Ocean, but in short supply.
  • Serves a variety of ecological roles, acting as a crucial habitat from micro-grazers all the way up to emperor penguins, which use the stable fast ice as a wintertime breeding platform.

What do we know about the formation, evolution, and decay of fast ice? 

We now have a satellite-derived dataset of 18 years of continuous observations of large-scale Antarctic fast ice extent (2000 to 2018). From this dataset, we can determine the timing of fast ice formation, maximum, decay, and minimum. However, the discrimination between pack and fast ice is still largely manual, i.e., requiring considerable time for a researcher to “map” the fast ice, thus reducing the objectivity of the maps. 

On a local scale, we have time series measurements of fast ice formation and decay from ice mass balance stations at a limited number of sites around the continent – however these tend to be found in “smooth” fast ice, meaning that regions of “rough” fast ice (i.e., forming when pack ice becomes fastened due to onshore winds) are poorly understood. Overall, fast ice observations are discontinuous, sparse, and under-described.

(a) Distribution of anomalies in Antarctic fast-ice extent in early-mid March 2022 (baseline: 2000 to 2021) with a value of –1.0 indicating a lack of fast ice in 2022. Inset (b) is a time series of early-mid March fast-ice extent from 2000 to 2022. Credit: Fraser and Wongpan et al. [2023], Figure 13

Why has most research to date focused on Arctic rather than Antarctic fast ice? 

A number of factors have converged to give us a much better understanding of Arctic fast ice, including:

  • Indigenous knowledge: fast ice has been used for transportation and hunting purposes for generations.
  • Charting: high-quality, weekly ice charts in the northern hemisphere have been recently used to map trends in fast ice extent.
  • Proximity: Vast tracts of Arctic fast ice are accessible from the northern coasts of Greenland, Russia, USA, and Canada, and it can also be found around Svalbard. By contrast, Antarctic fast ice requires expensive logistics and considerable planning for comprehensive research to take place.
  • Change: Arctic sea ice has been declining ever since the start of the modern satellite era (and likely earlier), whereas Antarctic sea ice was thought to be comparatively resilient against change (until 2015 when its variability increased).

What different techniques have been used to observe and measure Antarctic fast ice over the past century? 

The earliest accounts of Antarctic fast ice were from the Heroic Age of Exploration. Wright and Priestly (1922) accurately mapped the fast ice encountered on the Terra Nova expedition (1910-1913) and commented on the differences between “smooth” and “rough” fast ice. Nowadays fast ice is measured both in situ, and remotely (from both aircraft and spaceborne instruments).

Satellite imagery (left) and airborne photographs (right) of level and rough fast ice. Credit: Credit: Fraser and Wongpan et al. [2023], Figure 1

In situ measurements include ice mass balance observations, to determine the ways in which fast ice grows, as well as cores taken to measure the structure and chemical composition of fast ice. Remote measurements include satellite-based mapping, as well as measurements of its thickness, using altimetric and other techniques. Recently, ground-based radar interferometry for monitoring fast ice dynamics was developed and used for Alaska’s Arctic coast, but this hasn’t been used in Antarctica yet.  

How has remote sensing provided better data and new insights into fast ice distribution, seasonality, and thickness? 

Studies of the distribution of fast ice would be impossible without satellite remote sensing, due to its vast scale. Until now, the most complete large-scale dataset of Antarctic fast ice extent has come from visible/thermal infrared satellite imagery, but this imagery is heavily cloud-affected and requires manual interpretation for fast ice mapping.

More recently, satellite synthetic aperture radar (SAR) imagery has been used for fast ice mapping in both poles. SAR imagery is not affected by clouds or darkness, so is particularly suitable for this task. As we move toward automated fast ice mapping from SAR imagery, we can look forward to higher resolution, more accurate, and more timely fast ice maps. Remotely determining the thickness of fast ice is still a major challenge.

Innovative researchers have recently been using electromagnetic induction-based techniques to measure fast ice thickness from a low-flying aircraft suspending an electromagnetic induction device and flown just 15 meters above the surface, but this technique is impossible from space. 

How is fast ice in the Antarctic expected to change by the end of the 21st Century? 

The short answer is that we don’t really know. This is because climate models don’t yet simulate fast ice. It is hoped that within the next few years, fast ice will start to be included into many more models, giving us the ability to project fast ice extent and thickness into the future. Until then, what we can do is look at the changes in the ocean and the atmosphere projected by 2100, and guess what might happen to fast ice. Based on this, we think fast ice will decrease in extent, thickness, and duration, but we can’t quantify this.

What are some of the unresolved questions where additional research, data, or modeling are needed? 

We have very limited knowledge of fast ice thickness, other than in situ measurements which are very limited. As spaceborne altimeters become higher in resolution and more precise, we hope that this limitation becomes addressed soon, but more research is required.

Most of our current fast ice knowledge comes from smooth fast ice – but we don’t know much about rough fast ice, including its thickness, its storage of nutrients and its role in ecosystems. The problem is that most Antarctic stations are located near smooth, rather than rough, fast ice, so in situ measurements are difficult to achieve safely. 

In order to more fully understand the drivers of fast ice distribution, and its susceptibility to a changing climate, we need fast ice to be fully incorporated into our climate models. Without this vital tool, we cannot achieve the full understanding of Antarctic fast ice that is needed to fully assess its importance in the Earth system.

—Pat Wongpan (pat.wongpan@utas.edu.au, 0000-0002-7113-8221), Australian Antarctic Program Partnership, University of Tasmania, Australia; and Alexander D. Fraser (adfraser@utas.edu.au, 0000-0003-1924-0015), Australian Antarctic Program Partnership, University of Tasmania, Australia

Citation: Wongpan, P., and A. D. Fraser (2023), Landfast sea ice: the most important ice you’ve never heard of, Eos, 104, https://doi.org/10.1029/2023EO235023. Published on 22 August 2023.
This article does not represent the opinion of AGU, Eos, or any of its affiliates. It is solely the opinion of the author(s).
Text © 2023. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.