Polar bear researchers hiding significant increase in Southern Hudson Bay numbers

Last December, researchers vigorously promoted a possible 27% decline in Western Hudson Bay (WH) polar bear abundance but kept hidden the fact that adjacent Southern Hudson Bay (SH) numbers increased by 30% over the same period.

And surprise, surprise: the bombshell SH results call into question everything the ‘experts’ have been saying about polar bears in Hudson Bay for years.

I finally got a copy of the 2021 WH survey report from the Nunavut government, which was reported on by the media around the world in December 2022. The Nunavut government also sent along a copy of the 2021 SH report (helpfully asking, “would you also like the SH report?”), published at virtually the same time. The existence of a SH report was never mentioned by any of the media articles in December, even though it was referenced several times in the WH report, which suggests reporters never actually saw the WH report but were simply given a press release with approved talking points.

WH Report

The WH 2011 population estimate was 949 (range 618-1280), the 2016 estimate was 842 (range 562-1121), and 2021 estimate was 618 (range 385-852). The media must have been given the 27% figure representing the drop between 2016 and 2021 in a press release, because that figure isn’t mentioned in the WH report.

According to Andrew Derocher, of the University of Alberta, regarding the WH population, “The actual decline is a lot larger than I would have expected.” [CBC News, 23 December 2022]

I’ll say. However, the apparent decline from 2016 to 2021 was not statistically significant and neither was the decline between 2011 and 2021: the authors state this explicitly on pg. 29 of the WH report.

Oddly, that’s not the impression the press seems to have been given considering phrases used in their headlines, such as “population plummets,” “polar bears vanishing,” “polar bears…in sharp decline,” and “polar bears disappearing fast.”

The WH report authors state that abundance of adult male bears remained unchanged over the three sampling periods but that a decline in abundance of adult females and subadult bears (especially in the area around Churchill) seems to have driven an apparent decline in overall numbers. A suggestion was made that perhaps some Churchill-area females and subadults had simply relocated a bit north or south but the authors shot that notion down.

Although, the estimated abundance of adult female and subadult bears in Area 2 [i.e. around Churchill] decreased significantly between the 2011 and 2021 surveys, concurrent increases of these types of bears in Areas 1 (Cape Tatnum) [to the south] or 3 (Nunavut) [to the north] of WH were not found (Table 13). [WH report, Atkinson et al. 2022, pg. 32]

So, the missing females and subadults didn’t move a bit north or south, but did they perhaps move further south, into Southern Hudson Bay?

Notably, between 2016 and 2021, the estimated abundance of SH increased by 223 bears while that of WH decreased by 224 (Northrup et al. 2022). Changes in both subpopulations, at least between 2016 and 2021, could therefore be accounted for by movement of WH bears into SH. [WH report, Atkinson et al. 2022, pg. 24, my bold]

However, the WH report reveals that results of an unpublished SH biopsy study led by graduate student David McGeachy found 22% of bears formerly sampled in WH were found in SH in 2021 but these were predominantly adult males, not the “missing” adult females from WH. This means that a large movement of WH females and subadults into SH in 2021 cannot explain the simultaneous abundance decline in WH and increase in SH.

Our findings that adult female and subadult abundance has declined while adult male abundance has remained unchanged are thus inconsistent with a range-shift hypothesis. [WH report, Atkinson et al. 2022, pg. 33, my bold]

The authors suggest the apparent WH population decline was “consistent” with predictions of adverse effect of climate change on polar bear health and survival, yet provided no sea ice data for the relevant period. However, observations by others (including the authors of the SH report) indicate the decline couldn’t have been caused by poor sea ice conditions over the last five years because WH ice from 2017-2020 was better than it had been for decades.

In short, the authors of the WH report were unable to explain why there was an apparent decline in abundance between 2011 and 2021, except that several hundred adult females and subadults seemed to have disappeared from the central WH area around Churchill. The authors seemed unsure that an actual decline had taken place at all, given the uncharacteristically tenuous language used to describe their results:

Estimates derived for the WH subpopulation indicated a possible decline in total bear abundance between 2011 and 2021. [WH report, Atkinson et al. 2022, pg. 29, my bold]

SH Report

SH numbers went from 943 bears in 2012 (range 658-1350) to 780 in 2016 (range 590-1029) to up to a whopping 1119 in 2021 (range 860-1454), an increase of 30% over five years (see pgs. 29 and 42). Oddly, the authors don’t mention if this is a statistically significant increase or not, but it seems likely it is.

The report authors initially claim that it is “highly implausible” that an increase of this magnitude could be due to natural increases in birth rate or reduced mortality alone over such a short time period (pg. 29). However, their data also indicated that 35% of all SH bears they counted were yearlings or cubs-of-the-year (more than found in WH in 2021) and the authors finally concluded that a natural increase in numbers did happen in conjunction with good sea ice conditions from 2017 to 2020 (pg. 31), perhaps in conjunction with immigration of some bears from another subpopulation that they couldn’t verify.

Reconciling the Reports

The authors of both reports were left sputtering to explain their results. A loss of hundreds of WH adult females and subadults between 2016 and 2021 is not consistent with the prevailing hypothesis that lack of sea ice drives long-term declines in polar bear abundance in WH because sea ice conditions from 2017 to 2020 were better than they had been in decades (only 2021 was not as good). And besides, SH had similar sea ice conditions over the same period and polar bear numbers there increased, as shown in a copy of Figure 22 from the WH report, below.

Questions remain. Such as, if 22% of bears sampled with biopsy darts in SH in 2021 were males from WH, why didn’t the WH survey register this as a loss of adult males? That’s a real conundrum.

More importantly, what did happen to the hundreds of adult females and subadults between 2016 and 2021? Did they die of starvation for some unknown reason unrelated to summer sea ice conditions or did they go somewhere else? Is it possible for a population to grow by 30% in only five years? Are the population estimates the researchers have generated simply way off? The scientists can’t say.

Although at least one of the media articles from December 2022 suggested that a nebulous lack of ringed seals might be responsible for the apparent decline of WH bears, the WH report contains no mention of a possible dearth of ringed seals as a causal factor.

Oddly, there is also no mention of Foxe Basin (FB) to the north as a possible destination for the missing WH females and subadults. Foxe Basin bears haven’t be been surveyed since 2010 but at that time were doing very well with an estimated population size of 2580 (Stapleton et al. 2012). Who knows what’s happened since but FB bears mix with WH and SH bears on the ice of Hudson Bay over the winter and sea ice generally persists in Foxe Basin well into August, which means Foxe Basin might be a suitable refuge area for WH females and young bears looking for more predictable sea ice conditions, just as Franz Josef Land is for some Svalbard area bears.

Overall, the results of these two studies reveal that polar bear specialists have no idea what’s actually going on with Hudson Bay bears. Without sea ice to blame, they’ve got nothing. They simply can’t explain their results.

Polar bears may be dying for unknown reasons unrelated to summer sea ice in one region and reproducing like crazy next door, or they’re moving undetected between subpopulation boundaries. And if polar bears are indeed moving between Hudson Bay subpopulation boundaries, they are not acting like the discrete units required for management decision-making and reputable scientific research, and that’s a big problem for everyone.

No wonder polar bear specialists didn’t want the press to get wind of the SH report.


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. pdf here.

Northrup, J.M., Howe, E., Lunn, N., Middel, K., Obbard, M.E., Ross, T., Szor, G., Walton, L., and Ware, J. 2022. Southern Hudson Bay polar bear subpopulation aerial survey report. Final report to Ontario Ministry of Natural Resources, 29 November 2022, pdf here.

Stapleton, S., Peacock, E., Garshelis, D., and Atkinson, S. 2012. Foxe Basin polar bear aerial survey, 2009 and 2010, final report. Nunavut Wildlife Research Trust, Government of Nunavut, Igloolik. Available online, Pdf here.

Comments are closed.