Polar bears, sea ice and the limits of a single global number

Polar bears are often described as a single global population facing one uniform trend. Their biology is circumpolar, but their conservation status is assessed through 19 subpopulations distributed among five range states: Canada, Greenland, Norway, Russia and the United States. Some have recent abundance estimates; others have little or no current information. A global total cannot show those regional differences.

Why the ice matters

Sea ice is the platform from which polar bears hunt seals, especially ringed seals. Bears travel across it, find breathing holes and haul-out sites, and use it to reach productive feeding areas. When ice breaks up earlier in spring or forms later in autumn, many bears spend longer on land or in less productive parts of the marine environment. The effect depends on where a subpopulation lives and how local ice conditions change.

Longer fasting can reduce body condition and, in some populations, affect reproduction and cub survival. These links are supported by long-term regional studies, not by one observation that can be applied to every bear. The Arctic has several sea-ice regimes, and local prey, weather, denning habitat and human activity also shape outcomes. A bear seen on shore is not, by itself, proof of starvation or a population decline.

What the population table actually says

The specialist group’s 2024 report separates estimates from trend assessments. A subpopulation estimate may come from mark–recapture, aerial surveys, genetic sampling or demographic modelling. The methods do not all measure the same thing, and each estimate has its own confidence interval and date. The report also labels trends as increasing, stable, decreasing or unknown only where the evidence supports that judgement.

Some Arctic regions have too little information for a confident trend. “Unknown” is not another way to say stable, and an old estimate should not be presented as a count made this year. The often repeated global range of roughly 22,000 to 31,000 comes from a synthesis published in the 2023 specialist report; the uncertainty interval is part of the estimate, not a margin that can be ignored.

Nor does a stable or increasing estimate in one place disprove the risk from continuing sea-ice loss. Population counts respond over time, and surveys may lag behind environmental change. Conversely, projecting a future decline is not the same as documenting an identical decline today in every region. The saiga’s recent recovery is a useful reminder that species-wide labels can conceal very different regional histories, although the pressures on steppe antelopes are not interchangeable with those on Arctic bears.

Climate exposure is strong; local outcomes still differ

Arctic warming has reduced the area, thickness and persistence of sea ice. Satellite records document the physical change directly. Researchers then combine those records with bear movement, body condition, survival and reproduction to test how the habitat shift affects animals. The strength of evidence differs by region because long-term fieldwork is expensive and some areas are difficult to reach.

Polar bears can use foods on land and may alter their movements, but these options do not necessarily replace the energy gained from hunting seals on ice. Adaptation has limits when the habitat used for a specialised feeding strategy becomes less available. The key question is not whether a bear can eat something on shore, but whether the alternative supplies enough energy and supports survival and reproduction across years.

Management cannot substitute for a stable habitat

International cooperation has reduced unregulated hunting since the 1973 Agreement on the Conservation of Polar Bears. Range states monitor harvests, protect important denning areas and work with communities to reduce dangerous encounters. Those measures can help populations withstand local pressures, but they cannot restore lost sea ice. Climate mitigation and careful planning for Arctic development remain central to the species’ long-term prospects.

The accurate summary is therefore two-part: the 2023 global estimate is around 26,000 bears, with substantial uncertainty, while the 2024 specialist review shows that population knowledge and trends are uneven. Sea-ice loss is a serious long-term threat, but claims about present numbers must name their date, region and method. That is more informative than saying either that polar bears are all disappearing now or that one global estimate proves they are safe.