News detail view

At the beginning of the CONTRASTS expedition, RV Polarstern accompanied France’s new Tara Polar Station into the sea ice for its first expedition, marking the first time she led another ship into the ice – and a milestone for the French team on their way to their first Arctic drift (Polaris 1) in 2026/2027. At the same time, an aerial surveying campaign with the polar research plane Polar 6 launched from Station Nord (Greenland). In this regard, Polar 6 made two flyovers of one of the CONTRASTS ice stations while readings were simultaneously taken on the ice, by drone, and by helicopter (Figure 1). Moreover, Polar 6 made two further flyovers of the CONTRASTS region and its ice stations, gathering valuable data from the air.

The sea-ice regimes

One central element of the expedition: the repeated visits to three different sea-ice regimes. Each regime was surveyed four times over six weeks (Figure 2). Autonomous devices continued the time series while the members of the expedition team were busy elsewhere. The regimes varied in their age, provenance, drifting, and in terms of the local atmospheric and oceanographic conditions:

• Regime 1: seasonal ice in the marginal ice zone (MIZ), thin and very smooth. In the future, this will likely be the dominant regime in the Arctic. This year, it drifted surprisingly far to the east.
• Regime 2: predominantly two- to several-year-old ice that was formed in the Russian Arctic and is often characterised by sediment inclusions. Whereas this regime previously shaped the face of the Central Arctic, it now undergoes intensive melting before reaching Fram Strait.
• Regime 3: several-year-old ice from the regions north of Greenland and Canada. Given its prolonged exposure to various pressures, the most prominent deformation can be found in the oldest ice. Once commonly found, the percentage of this type of ice has now declined significantly.

The first analyses confirmed that the team had picked the right floes; they were able to determine the ice’s drift routes, and with them, the regimes’ provenance (Figure 3). At the beginning of the expedition, during the first visit to each regime, major differences (contrasts) became apparent: Regime 1 was heavily melted, riddled with meltponds, and porous. Regime 2 also showed surface melting, though less extensive. Regime 3 was largely intact at the outset but underwent increasing melting in the course of the expedition. Accordingly, the journey into the ice also became a journey through time: two weeks after the initial readings were taken, Regime 2 was in the same state as that first observed in Regime 1. And this trend continued. Generally speaking, the sea ice initially melted at the surface and later from below. As the patches of open water grew, it also began melting on the sides. In this regard, there were certain similarities between the three regimes’ melting behaviour. Figure 4 shows these developments from above and below, using images captured by the camera buoys.

Sea-ice situation in summer 2025

The sea ice in Fram Strait and in the region surveyed was exceptionally old compared to the last two decades – i.e., it was substantially older than the ice typically found there in past years. Normally, older ice is also thicker, yet the mean thickness of the undeformed, thermodynamically formed ice was only 1.5 metres. This is indicative of heavy melting, which was also reflected in the numerous patches of open water: roughly 15 percent of the helicopter- and airplane-based profiles were gathered over open water – an unusually high percentage for the region and time of year. In addition, the ice surveyed was subjected to a phase of extremely high temperatures during its winter drift through the Arctic. This may have further slowed ice growth and contributed to the relatively low thickness despite the high age. Further, the ice showed little deformation given its age and position: on average, we observed only 5 or 6 pressure ridges for every kilometre we flew over. As such, the negative trend of fewer ridges and a flatter surface topography continued – conditions that most likely accelerated the formation of meltponds throughout the campaign.

Originally, only three repeat measurements at each station were planned, as the team expected to progress through the sea ice on RV Polarstern at a mean pace of two to three knots. But due to the extremely loose ice conditions and extensive areas of open water, they could often move at five knots or more, greatly reducing the transit time between stations. The mean ice concentration in the survey region was at an all-time low (Figure 5). This was due to the extremely divergent drift, which was also responsible for the first station shifting far to the east.

Findings on the atmosphere, ocean and ecology

The atmospheric readings revealed considerable differences between regimes. The same weather conditions produced different reactions depending on the ice thickness, surface colour and melting status. Energy flows were greatly shaped by rain, snow, meltponds and surface structure (Figure 6). At an altitude of 300 metres, we measured air temperatures up to +13 °C – unusually warm for the region. Thanks to our extended stay in all three regimes, highly representative datasets on the synoptic conditions were collected.

We also found marked differences in the ocean: whereas sea ice covered colder and fresher water masses in the Last Ice Area (the region north of Greenland and the Canadian Arctic Archipelago), the open ocean served as a heat reservoir, even affecting the air masses above it. Thanks to measurements taken at depths of up to 1,000 metres, we were able to identify currents and eddies that significantly influence melting (Figure 7).

One surprise for the research team: throughout the expedition, little or no sea-ice algae was found in or below the floes – not even in the Last Ice Area, which was largely untouched by melting at the outset. Similar conditions were discovered two years earlier, during the ArcWatch 1 expedition, whereas previous expeditions had mostly reported extensive ice algae. Whether this is due to a dramatic decline of sea ice or to previous melting is currently being investigated on the basis of sediment samples gathered at a depth of 4,000 metres. With the aid of microscopes and the new planktoscope system, it was possible to determine that there were very few ice-algae cells among millions of phytoplankton. Instead of the algae, bacteria and ample zooplankton dominated the ecosystem and ensured that organic material was transported to the deep.

Technology: focus on autonomous systems

Another hallmark of CONTRASTS: the extensive and varied use of autonomous monitoring systems. Numerous buoys continuously observed and surveyed the melting of the sea ice. A particular highlight: new camera buoys, which documented the changes in images captured both above and below the sea ice (Figures 8 and 4). Accordingly, the ice could be monitored from both perspectives during the research team’s absence, which will greatly facilitate analysis of the datasets. While the tried and true ROV “Beast” surveyed the sea ice from below at all stations (Figure 9), ROVs with even more autonomous designs were also put to the test. As they can be remotely controlled “from home”, in the future they will allow spatial readings to be taken below the sea ice without the presence of research teams; to date, this has only been possible in the form of stationary devices deployed during campaigns. Further, the new autonomous surface vehicle “Otter” (Figure 10) passed its first field test in the sea ice during the CONTRASTS expedition. In the future, it will be used to take oceanographic readings in leads and in coastal and near-glacier settings. The first extensive measurements are slated for PS 150, the next expedition after CONSTRASTS.

Above the ice, the expedition team benefited from recent advances in drone technology. As a result, they were able to create high-resolution terrain models of the sea ice at all stations, produce additional horizontal and vertical air profiles, and map the surface albedo. Further, the new planktoscope offered them their first AI-assisted analytical tool that could be used to determine the composition and frequency of occurrence of the plankton community from on board – something that was previously next to impossible on a moving research vessel.

What happens next on Floe 3?

When they departed the floe for Regime 3, the CONTRASTS team left behind a selection of autonomous systems (Figure 11), which will continue gathering data and observe the transition to the freezing season. This part of the CONTRASTS programme will benefit from the subsequent expedition “East Greenland Current (EGC) Sources”, led by Prof Torsten Kanzow (AWI). As part of the ongoing expedition, Regime 3 will be visited again a few weeks from now. Then the last systems will be retrieved and certain readings will be taken once again for comparison. For this purpose, Marcel Nicolaus remained on board. In his words: “Four months on board is a long time, but also very exciting. Especially in these transitional periods between seasons, the sea ice changes so rapidly – and being able to experience it live is a real thrill.”

Meanwhile, the other expedition members are now on their way home or to other projects – because the real work begins, as always, once the expedition is over: now the wealth of samples collected will be prepared, analysed, and combined with the sensor data to find the answers to key questions concerning the melting in the various regimes. Someday, there may be a follow-up expedition – perhaps in a different season. It would be particularly valuable to take a closer look at the sea-ice formation processes in the autumn and winter.