If Steve Amstrup and his co-authors are right, the polar bears may be making their last stand in the Canadian high Arctic in 100 years: Polar Bears in the Southern Beaufort Sea II: Demography and Population Growth in Relation to Sea Ice Conditions.
Amstrup et al., working for the U.S. Geological Survey, grouped the world's 19 polar bear sub-populations into four general ecoregions - shown in the figure below (the two-letter codes identify the different subpopulations):
The light purple area is the "Polar Basin Divergent Ecoregion" where ice is formed and "advected" or moved away. The light green area is the more southerly Seasonal Ice Ecoregion. The forecast in both regions is for polar bear extirpation in 45 years.
The light blue area is the "polar Basin Convergent Ecoregion" where ice formed elsewhere tends to collect. (To get a clearer picture of the ice movements between the Polar Basin Divergent and Convergent Ecosystems, see Old Ice and the Northwest Passage) Polar bears are forecast to be extirpated here in 75 years.
Finally, the yellow area is the Archipelago Ecoregion. Here, polar bears may survive through the end 2100, but the population should be smaller.
The analysts experimented with the sensitivity of their models to changes in their sea ice coverage assumptions:
The modelers also used their model to see if policy interventions - other than those that might affect sea ice conditions - could change the results. We're talking here about things that people might be able to control, such as the size of quotas available for hunting, the level of oil and gas development, or measures to mitigate disturbance that might be associated with shipping.
The conclusions were not substantively affected in the Polar Divergent Ecosystem, or the Seasonal Ice Ecosystem:
...The conclusion for the Seasonal Ice and Polar Basin Divergent Ecoregions is that management of localized human activities can have no qualitative effect on the future of polar bears in the Seasonal Ice and Polar Basin Divergent Ecoregions if sea ice continues to decline as projected. Polar bears of both ecoregions are projected to move toward extinction by 45 years from now.
The marginal benefit for polar bears of investments in these activities (for example, reducing hunting, or steps to mitigate oil and gas development impacts on bears) is likely to be zero over the relevant range of available actions. Note: Sept 21: I came back to this topic and modified this pessimistic conclusion somewhat after spending more time with the paper: How likely is it that polar bears will become extinct?
Things were more hopeful for the Polar Basin Convergent Ecosystem and the Archipelago Ecosystem:
The conclusion from these fixed runs of the model is that management of human activities has the potential to qualitatively improve the welfare of polar bears in the Archipelago Ecoregion through the 21st century and in Polar Basin Convergent Ecoregion through mid-century....
That is, ultimately only mitigation measures in the Archipelago Ecosystem can have a long term effect. Measures in the Polar Basin Convergent Ecoregion help preserve polar bear populations for a while longer there. Different levels of protection actions in these areas may have non-zero marginal benefits to be matched against the marginal costs of taking the actions.
If the USGS scientists are right, by 2100, any remaining polar bears are likely to be a Canadian responsibility. Under current Canadian practice, polar bear management is primarily the responsibility of the provinces and territories, who share it with local Inuit. Almost all the remaining populations in 2100 fall within the Inuit majority territory of Nunavut. There is one population shared with Greenland (Kane Bay or "KB" on the map) and another, within Canada, that is divided between Nunavut and the Northwest Territories (Viscount Melville, or "VM" on the map).
The Archipelago populations are also exploited by local Inuit. There are several communities that fall within the ranges of these populations: Grise Fiord, Resolute, Arctic Bay, Kugaaruk, Talyioak, Gjoa Haven, and Cambridge Bay. The 2006 census says that about 5,100 people lived in these towns and villages. I haven't counted a few communities which are close to the area.
The 2005 population estimates by the Polar Bear Specialist Group of the International Union for the Conservation of Nature indicates that there are about 4,900 bears in these six populations. Half of these are in the Lancaster Sound (LS) population. Some of these other populations, where habitat is not very productive (not a lot of ringed seals) number less than 200. If I read the IUCN table correctly, the sum of the harvest quotas for these populations was about 178 bears. I assume these quotas will have to get much smaller or be eliminated.
This region may have a lot of new development. If the ice disappears, new shipping routes are possible, right through its center. There are plausible oil and gas prospects inside of it, and on its edges.
The Archipelago Region is shown in yellow in the figure. Here's a blow up of the yellow area with the channels of the Northwest Passage, and the location of the Sverdrop and Beaufort/Mackenzie oil and gas basins drawn in:
The red lines show the two main channels of the Northwest Passage. The northern blue circle is the Sverdrop oil and gas field, the southern blue circle is the Beaufort Sea/Mackenzie delta field. The map doesn't show anything about other types of development, such as the naval port the Canadians may develop at Nanisivik, the army training center at Resolute, and possible mining developments.
Eco-tourism is generally considered non-consumptive use. However, vessel traffic and infrastructure development associated with tourism, and the tourists themselves, may be a source of disturbance, and low level chronic pollution.