The official population estimates generated by the IUCN Polar Bear Specialist Group (PBSG) give the impression that the global total of polar bears has not changed appreciably since 2001:
2001 PBSG report 21,500-25,000

2005 PBSG report 20,000-25,000

2009 PBSG report 20,000-25,000

2013 PBSG website 20,000-25,000
However, some accounting changes were done between 2001 and 2009 (the latest report available) that mean a net increase in numbers had to have taken place (see summary map below and previous post here. Note: this is a different issue than the misleading PBSG website graphic discussed here).

And while it is true that population "estimates" are just that - rather broad estimates rather than precise counts - it is also true that nowhere do the PBSG explain how these dropped figures and other adjustments were accounted for in the estimated totals.

The simple details of these changes are laid out below, in as few words as I could manage, to help you understand how this was done and the magnitude of the effect. It's a short read - see what you think.

Polar bear subpopulations as defined by the PBSG: Top, in the 2001 report; Bottom, 2009 report. Map courtesy PBSG, with a few labels added and the subpopulations identified where “accounting” changes or adjustments to estimates took place.
SB, Southern Beaufort; NB, Northern Beaufort; VM, Viscount Melville; MC, M’Clintock Channel; LS, Lancaster Sound; GB, Gulf of Boothia; NW, Norwegian Bay; KB, Kane Basin; WH, Western Hudson Bay
Changes in the early 2000s

Between 2001 and 2005 the 'Queen Elizabeth Islands' region in northern-most Canada (see map above) was dropped altogether as a distinct subpopulation (along with its estimated 200 bears) and the tentative total of 2,000 bears estimated for 'East Greenland' was also dropped (from both minimum and maximum portions of the "21,500-25,000″ range).

In addition, about 1,000 was added to the minimum and about 1,000 subtracted from the maximum totals because a new, more accurate estimate for the Barents Sea became available in 2004, replacing the guess of "2,000-5,000" used in the 2001 report.1

That means between 2001 and 2005, due to accounting and 'upgrade' changes only, a total of 1,200 bears was removed from the minimum portion of the global estimate and 3,200 removed from the maximum portion of the global estimate, changes that had nothing to do with documented declines in subpopulations. These changes were all the result of differences in the way the data was presented (or not).

As a consequence, if the total population of polar bears had otherwise been unchanged or declined in the early 2000s, the total in 2005 should have been 20,300-21,800 - or less. However, the official total for 2005 was given as "20,000-25,000" - reflecting a reduction of 1,500 in the minimum portion of the estimate only (i.e. down from "21,500-25,000" in 2001).

This means there must have been a net increase of 300-3,200 bears (average 1,750) among the remaining subpopulations during the early 2000s. A comparison of the tables suggests these increases occurred in the Lancaster Sound, Gulf of Boothia and Davis Strait subpopulations.

Changes in the late 2000s

In 2009, the tentative total of 2,000 bears estimated for the 'Chukchi Sea' subpopulation was dropped - from both the minimum and maximum total estimates - because it was deemed not accurate enough: "2,000″ became "zero," even though bears are well known there (see previous posts here and here).

In addition, in the 2009 report, the Barents Sea subpopulation estimate was changed: given as "2,997 (2,299-4,116)" in 2005, based on studies completed in 2004, it was adjusted down to "2,650 (1,900-3,600)" in 2009, based on the same 2004 data. This 'adjustment' dropped another 350 bears from the minimum total2 and a whopping 500 bears from the maximum total (numbers rounded to keep things simple 2) that again, had nothing to do with a documented change in the number of bears in the Barents Sea but rather, a change in accounting.

But did the total estimate of polar bears worldwide in 2009 drop from "20,000-25,000" to "17,650-22,500" to reflect the removal of the Chukchi estimate and the adjustment in the Barents Sea figure? No, it did not.

[Now, the Barents Sea estimate is not the only subpopulation estimate that was 'adjusted' from one number to another based on the same survey data, but since we are tracking the Barents Sea figures from a "guess" in the 2001 report to a scientific estimate in 2009, it's important to include this final adjustment]

As a consequence, because the global estimate did not change between 2005 and 2009, there must have been another net increase of 2,350-2,500 polar bears among the remaining subpopulations in the late 2000s.

In summary, in order for the worldwide estimate of polar bears to have remained virtually unchanged since 2001, the global population must have increased by 2,650-5,700 bears (average 4,175) between 2001 and 2013. These increases did not off-set the slight declines in other subpopulations, as the unchanging totals imply, but were in addition to them.

In other words, it appears that the global population of polar bears could not have remained stable since 2001 - it had to have increased by an average of almost 4,200 bears!


1. The Barents Sea subpopulation was roughly estimated to have been "2,000-5000" in 2001. This means that of the "20,000-25,000" total estimate, 2,000 of the "20,000" minimum came from the Barents Sea and 5,000 of the "25,000" maximum came from the Barents Sea. However, the contribution of Barents Sea bears to the total changed in 2005 because a new, more accurate estimate became available (stated as "2,997 average, range of 2,299-4,116"). No other subpopulation estimate, except the Barents Sea, changed from a tentative, 'educated guess' to a science-based estimate between 2001 and 2012, so I contend it is valid to include this as an "accounting change."

2. In 2005, as far as I can tell 2,997 of the "20,000" total minimum estimate came from the Barents Sea, and 4,116 of the "25,000" maximum estimate came from the Barents Sea (that's because in contrast to previous years, in 2005 the total estimate range (i.e. "20,000-25,000") appears to be the total of the average estimate for each subpopulation (rather than the minimum) and the total of the maximum estimate for each subpopulation, see previous post here.


2001 PBSG report:
Lunn, N.J., Schliebe, S., and Born, E.W. (eds.). 2002. Polar Bears: Proceedings of the 13th working meeting of the IUCN/SSC Polar Bear Specialists Group, 23-28 June , 2001, Nuuk, Greenland. Gland, Switzerland and Cambridge UK, IUCN.

2005 PBSG report:
Aars, J., Lunn, N. J. and Derocher, A.E. (eds.) 2006. Polar Bears: Proceedings of the 14th Working Meeting of the IUCN/SSC Polar Bear Specialist Group, 20-24 June 2005, Seattle, Washington, USA. Occasional Paper of the IUCN Species Survival Commission 32. IUCN, Gland, Switzerland and Cambridge, UK.

2009 PBSG report:
Obbard, M.E., Theimann, G.W., Peacock, E. and DeBryn, T.D. (eds.) 2010. Polar Bears: Proceedings of the 15th meeting of the Polar Bear Specialists Group IUCN/SSC, 29 June-3 July, 2009, Copenhagen, Denmark. Gland, Switzerland and Cambridge, UK.