Share your science:

IceBird view of the Norwegian R/V Kronprins Haakon en route to collect sea-ice thickness data at the location where scientists onboard the ship took in situ ice measurements.

Arctic sea-ice monitoring: from the Vikings to satellite observations

SHARE YOUR SCIENCE: The first Arctic sea-ice observations were documented by the Vikings, and later followed by the first polar explorers and sailors. Today, the IceBird programme is part of an international monitoring series, documenting the Arctic ice cover as it undergoes rapid changes due to climate warming.

Published

The Arctic Ocean spans roughly 14 million square kilometres - about half the area of Africa – and is, at least partly, covered by sea-ice year-round. This ice cover plays a fundamental role in the Earth’s climate system. It influences the global ocean circulation, prevents energy absorption from the Sun due to its high albedo, and limits heat exchange between the ocean and the overlying atmosphere.

Changes in the sea-ice cover can therefore have implications reaching far beyond the Arctic region. Observing these changes requires systematic monitoring of the Arctic Ocean, which is difficult due to its vastness, remoteness, and harsh environment.

The long history of sea-ice observations

The oldest record of sea-ice observations dates back to as early as 870 A.D. and was assembled by the Vikings. The record contains the number of weeks per year that sea-ice was seen along the north coast of Iceland. It was not until the 15th-16th century that other sea-ice records showed up, this time collected by the first polar explorers and sailors such as John Cabot and Willem Barentsz, in their quest for Arctic shipping routes.

Print by Jan Luyken showing walrus hunting during Willem Barentsz' expedition to Novaya Zemlya in 1596.

The notes from the polar explorers, and later whalers and trappers, provide insightful information on sea-ice conditions but are scarce both in time and space. The first actual sea-ice thickness measurements were taken during the Jeannette Expedition from 1879 to 1881 using an auger. 

As aviation picked up in the early 20th century, it allowed for observations across larger spatial scales, such as during the first North Pole overflight by Roald Amundsen in 1926.

The second half of the 20th century brought new technologies that enabled us to monitor Arctic sea-ice in a systematic manner for the first time. Submarines equipped with upward-looking sonars provided estimates of sea-ice draft along their tracks.

Since the late 70’s, microwave radiometers onboard satellites have provided continuous sea-ice observations across the Arctic, leading to the first maps showing sea-ice area and extent, as well as routinely produced ice charts for navigation purposes.

Nowadays, satellites carry a variety of sensors acquiring sea-ice observations, such as altimeters providing estimates of freeboard height that are subsequently converted to sea-ice thickness estimates, or synthetic aperture radar (SAR) sensors providing imagery of the sea-ice cover which can be used to estimate ice drift.

The Norge airship floating above Svalbard before its departure to the North Pole in 1926

We still need to be on the ice

Even though the spaceborne observations give us the spatiotemporal coverage needed to monitor the Arctic Ocean and its sea ice, we still rely on in situ ('on the ice') or airborne measurements to validate the information retrieved from satellite data. Dedicated field campaigns collect sea-ice observations that are used as comparison material.

Nevertheless, the large discrepancies in spatial and temporal scales make it hard to directly compare the different observations: from instant point measurements such as drill-holes or snow pits, to airborne measurements with drones or planes that cover tens to hundreds of kilometres over days, to satellite observations spanning the entire Arctic year-round.

The German Alfred Wegener Institute, Helmholtz-Center for Polar- and Marine Research (AWI), has been running sea-ice surface roughness surveys since 1993 and sea-ice thickness surveys since 2003. The IceBird programme is part of this monitoring series in the form of airborne sea-ice thickness surveys run twice a year, in summer and winter.

These campaigns have documented, in detail, the Arctic ice cover as it has changed rapidly due to climate warming over the past few decades. More recently, the airborne measurements offer a unique reference for validating satellite observations of sea-ice. During the survey flights, sea-ice thickness is inferred from a combination of electromagnetic (EM) induction measurements and laser measurements.

The induction measurements are obtained with the ‘EM-Bird,’ a torpedo-like sensor towed under the airplane at an altitude of 15 metres above the sea-ice. This provides the distance between the EM-Bird and the interface between the ice and underlying sea water. 

The corresponding laser measurements provide the distance from the EM-Bird to the top of the snow or ice surface. The difference between the ice-water interface distance and snow or ice surface distance is the total sea-ice thickness, as shown in the illustration.

Sea-ice thickness retrievals using electromagnetic Inducüon and laser measurements.

In winter, this is rather the combined snow and ice thickness, whereas in summer, it is just the ice thickness, as the snow has largely melted.

The summer 2024 campaign

The summer 2024 campaign is operating from Station Nord in Greenland and carrying out survey flights over the Lincoln Sea and Fram Strait. Apart from the usual monitoring transects, this year’s campaign collected sea-ice thickness data over an in situ sea-ice station of the Norwegian icebreaker R/V Kronprins Haakon at 86°N.

This will allow us to place the in situ measurements taken by scientists onboard the ship into a wider context. If the weather allows, we aim to survey a ‘satellite underflight,’ which is a flight over the ground track of an overpassing satellite altimeter.

The IceBird summer 2024 team. Back row from left to right: Gerit Birnbaum (AWI), Triston MacLean (Kenn Borek Air/K8A), Dwayne Bailey (KBA) and Jack Landy (UiT). Front row from left: Alan Gilbertson (KBA), Eduard Gebhard (AWI) and Catherine Taelman (UiT)

As we are surveying drifting sea-ice, it is important to minimise the time gap between the satellite overpass and the flight in order to acquire data over the exact same ice. 

Such a data set provides the opportunity to, for example, further investigate biases in sea-ice thickness retrievals from space. The AWI IceBird campaigns build upon a long legacy of sea-ice surveying in the Arctic, thereby providing essential data for climate studies.

(This article was originally published on Seaice Portal)

Share your science or have an opinion in the Researchers' zone

The ScienceNorway Researchers' zone consists of opinions, blogs and popular science pieces written by researchers and scientists from or based in Norway.
Want to contribute? Send us an email!

Powered by Labrador CMS