Even though no new record low was reached this year, the sea-ice extent consistently lies at the lower end of the range of values observed between 1981 and 2010 (Figure 1).
Based on the monthly mean value to date, from the first 17 days of September, the derived preliminary value is 4.43 (±0.05) million square kilometres, which would be the third-lowest value since 1979 (Figure 2; Figure 11). However, that number could still change by the end of the month. Generally speaking, the negative trend in summer sea-ice extent has continued in this year, with corresponding effects not just on the Arctic and its ecosystem, but also on global weather patterns.
In the following, we summarise the processes that influenced the sea ice in the past year and review how well the model-based predictions of the sea-ice minimum in September 2024, made since this June, match the reality. In addition, we provide a brief overview of past and ongoing expeditions, during which sea-ice measurements will be gathered this summer.
Satellite data shows declining thickness of very old ice
The summer melting period was preceded by a relatively mild winter. In January and February 2024, exceptionally high temperatures at 925 hPa (ca. 750 m above the surface) influenced the entire Arctic (Figure 3). This was followed by unusually high temperatures near Canada and Greenland, whereas lower temperatures were prevalent along the coast of Siberia (Figure 4). Due to the warm winter, thermodynamic models show considerably lower ice growth, especially in those regions characterised by comparatively old and thick ice to the north of the Canadian Arctic Archipelago. Ice-thickness readings taken by the CryoSat-2/SMOS satellites indicate pronounced negative anomalies in these regions (Figure 5). Thicker-than-average ice was observed by CryoSat-2/SMOS in the Eurasian Arctic, which could be due both to increased ice growth in March/April thanks to lower temperatures (cf. Figure 4) and to powerful deformation processes caused by ice convergence. The comparatively low level of export (Figure 6) of newly formed ice from the Russian shelf seas to the Central Arctic and Fram Strait was also unusual.
Summer melting
With the start of spring, temperatures started climbing throughout the Arctic and the ice began melting, initially at a rate comparable to the mean for the past ten years (cf. Figure 1). Atmospheric conditions were variable and differed from region to region. In June, a stable high-pressure cell formed in the Central Arctic, bringing with it cloudless skies and increased heat input (Figure 7). As a result, sea-ice retreat accelerated considerably in the summer months, and by the end of July there was comparatively little ice remaining along the coast of Siberia and in the Beaufort Sea (Figure 8). It was only in the East Siberian Sea (Wrangel Island) that, just as in the previous three years (2021 – 2023), the ice remained for a relatively long time, as a result of which the Northeast Passage wasn’t nearly ice-free until mid-August; at the same time, the Northwest Passage was already navigable. In August, temperatures and windspeeds were slightly below average, except in the Kara Sea and Wandel Sea. The moderate drift speeds and low temperatures in August may have at least limited sea-ice retreat in August and the first week of September. However, this year’s satellite-based surface temperature readings once again indicate significantly higher values in the marginal seas – a phenomenon now seen every year (Figure 9).
Predictions closely match the reality
In the past, AWI experts have used a range of methods to predict the September sea-ice extent several months in advance. This year, for the first time they applied a coupled atmosphere/ocean/sea-ice model, the AWI Coupled Prediction System. After the construction of initial conditions with the latest observational data, forecasts were launched at the beginning of each month from June to September. Each ensemble prediction consists of 30 individual simulations to reflect uncertainties. The prediction started on 1 June tended to overestimate the late-summer sea-ice extent (Figure 10). In particular, the rapid decline in extent in July and August was underestimated; specifically, a high probability of sea ice was forecast on 7 September in and near the Beaufort Sea and East Siberian Sea. But the prediction started on 1 July provided a more accurate picture of the decline and, as expected, subsequent predictions were increasingly accurate and precise.
On-site measurements in 2024
This year, too, the satellite- and model-based analyses are being complemented by on-site observations. For example, extensive ice-thickness and surface measurements are being taken near the research icebreaker Polarstern, which is currently in the Central Arctic. The ice there, just like last year, mostly originated in the Western Arctic. The thickness of level ice without pressure ridges is currently ca. 1.3 metres. By comparison, in previous years smaller thicknesses (ca. 0.8 to 1.0 metre) were observed in the same region. However, back then, the ice chiefly came from the Eastern Arctic and was, in some cases, exposed to higher temperatures in summer. On the basis of these observed differences, the researchers are now working to gain a better understanding of how different source areas of sea ice affect the ecosystem and summer melting.
In addition to gathering ship-based measurements, this year we once again deployed research aircraft. In late winter, the research aircraft Polar 5 was used to take extensive surveys in the Canadian Arctic and Fram Strait (contact: Christian Haas (AWI)). In addition to ice thickness, the surveys provide information on the depth of the snow cover, a parameter that still can’t be satisfactorily measured by satellite. In the summer months, there was a second airborne campaign north of Greenland and in Fram Strait (contact: Jack Landy (Arctic University of Norway)). The primary goal was to gather data that could improve the quality of satellite-based ice-thickness estimates in summer.
Contact
Questions?
Write us an email or use the contact form.
Graphics
