A surface that will not hold still
In the Bay of Bothnia, winter does not arrive as a single event. It arrives as a sequence of small decisions made by water: a skim of new ice in a sheltered bay, a night of clear sky and no wind, a morning when the surface has closed. The Finnish Ice Service’s own photograph of a sea-ice field and a ship track, taken from the bridge of the icebreaker Urho, shows what that surface becomes to the people who work it: not a backdrop but a document. The track is a line drawn across a page that is being rewritten underneath it.
This is the condition the Finnish word ahtojää names. In the operational literature of the Baltic, drift ice is described as having a dynamic nature, forced by winds and currents. Its motion produces an uneven, broken field: floes up to several kilometres in diameter, leads and cracks, brash ice barriers, rafted ice, and ridges. The field is never still. A chart of it is therefore not a map of a thing but a reading of a process, and the process does not pause while the reading is made.
The chart as a daily form
The Finnish Ice Service’s predecessor began operational ice charting in 1915 to provide ice information for winter navigation, according to the service’s own account published in Frontiers in Marine Science in 2025. The service typically operates from mid-October to the end of May. In recent years the start has been fixed to mid-October, when analysts begin monitoring the cooling of the sea surface. Daily chart production is typically triggered in mid-December, though it may begin as early as late November or as late as early January, when the icebreaking authorities issue assistance restrictions and the first icebreakers are sent out.
The daily rhythm is precise. Synthetic Aperture Radar satellite images, the main source of spatial information, are acquired in early morning and early evening orbits and are available to the analyst well before noon. The ice report is written before noon; the assisting analyst translates it to Swedish and English and publishes it around noon. The ice chart is finalized and published in the afternoon around 2 pm. Ice codes are generated at the end of the day. Darkness and cloudiness frequently prevent the use of optical satellite imagery until March, so for much of the winter the sea is read through radar and through the reports of icebreakers and coastal observers.
There is a structural resemblance here that I want to name carefully, because it is easy to overstate. A feed converts a moving world into a sequence of dated updates, each one superseding the last. The winter chart does something similar with a surface that cannot be assumed. The comparison is not causal and not metaphorical in any strong sense; it is a resemblance between two genres of attention. Both are attempts to hold a changing thing in a form that can be acted on before it changes again. The chart is the older form, and it is the one that has to be right.
Why the ice matters to the economy of a coast
The stakes are not aesthetic. Approximately 95% of Finland’s international trade is transported by sea, according to the same account, which cites Finnish Customs data. Finland and Estonia are the only countries in the world where all mainland harbours freeze during normal winters. Winter navigation is managed by the Finnish Transport Infrastructure Agency with support from Fintraffic VTS, the Border Guard, the pilotage company Finnpilot, and the Ice Service. The chart is not a scientific ornament; it is part of the infrastructure that keeps a country supplied through a season when its ports are closed by ice.
The ice season can last up to 220 days in the northern Bay of Bothnia. In the southern Baltic, along the German and Polish coasts and in the Danish Straits, the annual maximum level ice thickness seldom exceeds half a metre. The range across the basin is enormous, and the chart has to describe all of it, every day, in a form that a ship’s master and an icebreaker captain can use.
The season that no longer keeps its shape
HELCOM’s State of the Baltic Sea assessment states that the effects of climate change are evident in the Baltic Sea today: water temperature is rising, ice extent is decreasing, and annual mean precipitation is increasing over the northern part of the region. The European Environment Agency publishes a time series of maximum ice extent in the Baltic Sea covering the winters from 1719/20 to 2024/25, sourced from the Finnish Meteorological Institute and updated from the work of Seinä and Palosuo (1996) and Seinä et al. (2001). The series shows variability across three centuries; the fifteen-year moving average is the line to watch, and it is the line that has been trending downward in recent decades.
What the retrieved sources do not give me is a single number for how much the season has shortened, or a date by which a particular contraction occurred. I am not going to invent one. The documented facts are these: the ice cover usually begins to form in November and reaches its largest extent between January and March; normal break-up starts in April and the ice melts completely by the beginning of June; the service operates until the ice has practically melted from the northern Baltic, typically mid-May, varying from early May to early June. Those are the boundaries of a season. HELCOM says the extent is decreasing. The EEA series shows the long record. The precise rate of change is a question for the ice records themselves, and the records are open.
The labour question
Here is the argument I want to make, and it is an interpretation rather than a finding. The shortening ice season is not only an ecological fact. It is a labour question. A routine built around a season that reliably ran from October to May is being asked to serve a season that no longer keeps that shape. The analysts still begin in mid-October, monitoring the cooling water. The daily charting still begins when the icebreakers go out. The service still continues until the ice has melted. But the middle of the season — the long, cold, stable stretch when the chart described a field that stayed roughly where it was put — is the part that is changing. The edges of the season are where the work is most intense and most uncertain, and the edges are where the change is felt first.
I want to be careful here. The retrieved sources do not state that the Finnish Ice Service has changed its operational calendar in response to warming. They describe the calendar as it is. What they do support is the observation that the calendar is built around a set of expectations — freeze-up in November, maximum extent in January to March, break-up in April, melt by June — and that HELCOM reports the ice extent is decreasing. The expectations and the trend are in tension. How that tension is managed operationally is a question the public record does not yet answer, and I will not pretend it does.
What the chart holds
A chart is a compression of knowledge. It holds the analyst’s reading of a radar image, the icebreaker captain’s report of where a vessel got stuck, the coastal observer’s note of a lead opening, the model’s forecast of drift. It holds, too, the accumulated judgment of a service that has been doing this since 1915: which signatures in a radar image mean ridged ice and which mean brash, how a wind shift will move a field, where a track can be trusted and where it cannot. That judgment is not written down in a single document. It lives in the routine, in the daily production of the chart, in the handover between analysts, in the co-production with the Swedish Ice Service that has run on a week-to-week basis since the winter of 2017/18.
If the ice season contracts, the routine does not disappear. But the conditions it was built to describe become less frequent, and the knowledge that accumulates through repetition becomes harder to renew. A chart of a mild winter is still a chart, and the analyst still reads it. But the mild winter teaches different lessons than the severe one, and the archive of severe winters grows older while the archive of mild ones grows longer.
An open question
The sea’s winter is a material condition that has shaped a coast’s economy, its labour, and its forms of attention. The chart is one of the ways that condition has been made legible. As the season shortens, the chart continues to be produced, and the question is not whether the routine will survive but what it will be for. If the ice season contracts, what happens to the knowledge held in the charting routine — and to the winter navigation it was built to serve?
Sources
- Eriksson, P. B., Vainio, J., Tollman, N., Jokiniemi, A., Arola, A., Mäkynen, M., Karvonen, J., and Kangas, A. (2025). The Finnish Ice Service, its sea-ice monitoring of the Baltic Sea and operational concept. Frontiers in Marine Science 12. https://doi.org/10.3389/fmars.2025.1561461
- HELCOM. State of the Baltic Sea 2023 — Climate change findings. https://stateofthebalticsea.helcom.fi/findings/pressures/climate-change/
- European Environment Agency. Maximum extent of ice cover in the Baltic Sea, winters 1719/20–2024/25. Source: Jouni Vainio, Finnish Meteorological Institute, updated from Seinä and Palosuo (1996) and Seinä et al. (2001). https://www.eea.europa.eu/en/analysis/indicators/arctic-and-baltic-sea-ice/maximum-extent-of-ice