From The polarbearscience

In my State of the Polar Bear 2023 report for the Global Warming Policy Foundation, I discuss recent news relevant to polar bear conservation and science issues. The most startling of these is the revelation that Western Hudson Bay polar bear numbers have not declined since 2004.

Press release (27 February 2024), Global Warming Policy Foundation

50 years after hunting ban polar bears are thriving, new report shows

London, 27 February: 2023 marked 50 years of international cooperation to protect polar bears across the Arctic. Those efforts have been a conservation success story: from a population estimated at about 12,000 bears in the late 1960s, numbers have almost tripled, to just over 32,000 in 2023.

Despite this dramatic increase in polar bear populations, claims that their numbers are falling due to climate change still dominate most media coverage.

Since 2004 we have been told that polar bear numbers in Western Hudson Bay have been steadily declining, but a new study made public in 2023 reveals that this isn’t actually true. In the State of the Polar Bear Report 2023, published by the Global Warming Policy Foundation (GWPF) on International Polar Bear Day, zoologist Dr. Susan Crockford provides the details on this explosive news.

Among other issues addressed in this year’s report, Crockford explains that population surveys of Western Hudson Bay polar bears completed in 2011, 2016, and 2021 generated lower estimates than a survey done in 2004. However, these differences in bear counts are not statistically significant from each other, which means there has been no negative trend during the last 20 years.

Crockford said the consequences of this survey are enormous.

“It means there has been no statistically significant trend in Western Hudson Bay polar bear numbers since at least 2004. This result absolutely guts computer model predictions of future polar bear catastrophe that assume Western Hudson Bay numbers have been steadily declining.” 

She also said the Western Hudson Bay survey raised questions about previously-unknown movements of hundreds of bears across subpopulation boundaries in Hudson Bay.

“There is now fairly strong evidence that the subpopulation boundaries in Hudson Bay need to be adjusted and have their previous population estimates recalculated. A similar issue is acknowledged in Alaska, where for decades fairly large numbers of Southern Beaufort polar bears have moved back and forth over the current boundary between the Chukchi Sea to the west and the Northern Beaufort Sea to the east. This means that if you count Southern Beaufort bears in one area at one point in time, it may look like a population decline has taken place when it actually hasn’t.”

Crockford added, “Western Hudson Bay and the Southern Beaufort are the only polar bear subpopulations which appeared to show strong support for the premise that sea ice declines blamed on human-caused global warming have reduced bear abundance. Now we know previous claims are almost certainly incorrect.”

Key Findings

There were no reports from the Arctic in 2023 indicating polar bears were being harmed due to lack of summer sea ice habitat, in part because Arctic sea ice in summer has not declined since 2007.

Contrary to expectations, a study in Svalbard found a decrease in polar bears killed in defense of life or property over the last 40 years, despite profound declines in sea ice over the last two decades.

A survey of Southern Hudson Bay polar bears in 2021 showed an astonishing 30% increase over five years, which adds another 223 bears to the global total.

A concurrent survey of Western Hudson Bay polar bears in 2021 showed that numbers had not declined since 2011, which also means they have not declined since 2004. Movement of bears across boundaries with neighbouring subpopulations may account for the appearance of a decline, when none actually occurred.

The IUCN Polar Bear Specialist Group has ignored a 2016 recommendation that the boundaries of three Hudson Bay subpopulations (Western HB, Southern HB, and Foxe Basin) be adjusted to account for genetic distinctiveness of bears inhabiting the Hudson Bay region; a similar boundary issue in the western Arctic between the Chukchi Sea, and the Southern and Northern Beaufort subpopulations, based on known movements of bears between regions, has been acknowledged since 2014 but has not yet been resolved.

The US Fish and Wildlife Service and the IUCN Polar Bear Specialist Group, in their 2023 reports, failed to officially acknowledge the new-found South-East Greenland population as the 20th subpopulation, despite undisputed evidence this is a genetically distinct and geographically isolated group. Numbers are estimated at 234 individuals.

Citation for the report: Crockford, S.J. 2024. State of the Polar Bear 2023. Briefing Paper 67. Global Warming Policy Foundation, London. Download pdf here.

Backgrounder: Western Hudson Bay population surveys

Here is a short summary of Western Hudson Bay (WH) polar bear population size survey results since 1987, which includes the 95% confidence interval calculated (listed as the ‘range’ of possible accurate estimates) for each mean estimate. These surveys used a dizzying mix of field methods, geographic areas covered within the WH subpopulation boundary (see map below), and mathematical models to generate population estimates.

1987, based on geographically-restricted (Area C only) capture-recapture study: 1184 (range 993-1411), calculated by Lunn et al. (2016:1313) for comparison to their 2011a study; note this figure is slightly different (i.e., 1194, range 1020-1368) than used by Regehr et al. (2007) to compare to 2004 data.

2004, based on geographically-restricted (Area C only) capture-recapture study: 935 (range 794-1076), a statistically insignificant decline of 22% since 1987 (Regehr et al. 2007:2679).

2011a, based on geographically-restricted capture-recapture (Area C + some of Area B): 806 (653-984) is the one most often quoted (Lunn et al. 2013, 2016). As happened in 2022 regarding the 2021 survey, in 2013 polar bear researchers shared results of the 2011 survey with the media before the report was available to the public.

2011b, based on comprehensive aerial survey: 1030 (range 754-1406), accepted by the PBSG in 2014 (Stapleton et al. 2014; see also PBSG 2023).

2011c, 949 (range 618-1280) is a recalculation of the 2011b estimate to make it comparable to 2016 and 2021 (see Dyck et al. 2017:3).

2016, based on an aerial survey with some differences from 2011b aerial survey: 842 (range 562-1121) (Dyck et al. 2017)

2021, based on an aerial survey with some differences from 2011b aerial survey: 618 (range 385-852), not statistically different from the 2016 or 2011a estimates. Atkinson et al. 2022:29 stated (my bold): Estimates derived for the WH subpopulation indicate a possible decline in total bear abundance between 2011 and 2021. …Although differences amongst these estimates [for 2011, 2016, and 2021] were not statistically significant, total abundance has declined consistently between successive surveys.

They say that “total abundance has declined consistently between successive surveys” as if that over-rules the previous statement but it does not. The rules of mathematics say that without statistical significance, there has been no decline over time, as I explain below.

Confidence intervals for W. Hudson Bay population size estimates, 1987-2021 (not to scale). Red intervals are based on limited-area capture-recapture data, blue from limited-area aerial surveys, and black from a full-subpopulation aerial survey. Dates of surveys are on the right, with mean estimates in brackets. Only non-overlapping intervals are statistically significant (** and *), see explanation below.

With two exceptions, all WH population size estimates since 1987 are not statistically different from each other because of the overlap in their ‘confidence intervals’ (i.e., the ‘range’ of possible correct estimates (e.g., 993-1411 for 1987). As shown in the chart above, the first and last survey years (1987 and 2021) have clear non-overlapping confidence intervals (marked **), with the caveat that entirely different methods were used for these estimates, so the we should be careful concluding that a real decline in abundance has taken place.

And while the 2011a estimate has a non-overlapping confidence interval compared to 1987 (marked *), indicating a statistically significant decline between 1987 and 2011, the interval does overlap with the estimate for 2004 and for every other estimate since then. This suggests that no definitive declining trend in WH polar bear abundance can be said to have occurred between 2004 and 2021.

Therefore, the comprehensive aerial survey (2011b) which generated an estimate of 1030 bears is likely still the best assessment of abundance — and since it is also not statistically different from the 1987 estimate, it suggests the WH population has probably been stable for more than three decades.

Polar bear researchers continue to make public statements claiming that a continuous population decline has occurred in WH when the data they have collected simply do not support that conclusion. They defend this falsehood by insisting that numbers of adult females and subadults have declined significantly the last 20 years, just as they predicted would happen due to declining sea ice.

But according to their own data, sea ice conditions have not declined over the last two decades. There was a step-change in breakup and freeze-up dates that happened about 1998, which in most years added about three weeks to the length of time that WH polar bears had to remain onshore. Since then there has been no change in the length of the onshore period, which they freely admit (Atkinson et al. 2022; Castro de la Guardia et al. 2017; Lunn et al. 2013, 2016; Miller et al. 2022; Stapleton et al. 2014).

And if lack of sea ice isn’t driving the apparent decline in adult females and subadult, what is? Movement between subpopulation boundaries, especially into Foxe Basin, is one strong possibility that has not been adequately addressed. See the polar bear report for details.

Atkinson, S.N., Boulanger, J., Campbell, M., Trim, V. Ware, J., and Roberto-Charron, A. 2022. 2021 Aerial survey of the Western Hudson Bay polar bear subpopulation. Final report to the Government of Nunavut, 16 November 2022.

Castro de la Guardia, L., Myers, P.G., Derocher, A.E., Lunn, N.J., Terwisscha van Scheltinga, A.D. 2017. Sea ice cycle in western Hudson Bay, Canada, from a polar bear perspective. Marine Ecology Progress Series 564: 225–233. http://www.int-res.com/abstracts/meps/v564/p225-233/

Dyck, M., Campbell, M., Lee, D., Boulanger, J. and Hedman, D. 2017. Aerial survey of the Western Hudson Bay polar bear sub-population 2016, final report (26 June 2017). Status report 2017-xx, Nunavut Department of Environment, Wildlife Research Section. Igloolik, Nunavut. pdf here.

Lunn, N.J., Stirling, I., Andriashek, D. and Kolenosky, G.B. 1997. Re-estimating the size of the polar bear population in Western Hudson Bay. Arctic 50(3): 234-240.

Lunn, N.J., Regehr, E.V., Servanty, S., Converse, S., Richardson, E. and Stirling, I. 2013. Demography and population assessment of polar bears in Western Hudson Bay, Canada. Environment Canada Research Report. 26 November 2013. PDF HERE

Miller, E.N., Lunn, N.J., McGeachy, D., and Derocher, A.E. 2022. Autumn migration phenology of polar bears (Ursus maritimus) in Hudson Bay, Canada. Polar Biology 45:1023-1034.

PBSG. 2023. ‘Status Report on the World’s Polar Bear Subpopulations’. IUCN Polar Bear Specialist Group, 17 October. https://www.iucn-pbsg.org/

Regehr, E.V., Lunn, N.J., Amstrup, S.C., and Stirling, I. 2007. Effects of earlier sea ice breakup on survival and population size of polar bears in western Hudson Bay. Journal of Wildlife Management 71(8):2673-2683.

Stapleton S., Atkinson, S., Hedman, D., and Garshelis, D. 2014. Revisiting Western Hudson Bay: using aerial surveys to update polar bear abundance in a sentinel population. Biological Conservation 170:38-47. http://www.sciencedirect.com/science/article/pii/S0006320713004618#