A newly-published paper by Martyn Obbard and colleagues in the journal Arctic Science claims a 17% decline in abundance of polar bears in the Southern Hudson Bay region after years of reduced sea ice and declining body condition (Obbard et al. 2018). The decline in numbers was not statistically significant but an additional statistical analysis (“Monte Carlo simulation”) not applied to any other estimate in recent years suggested the decline could be real, so a real decline is what was reported to the press.
Only one Canadian Press story has so far been circulated amongst outlets in the media (published hours after the paper appeared online, not at the same time), suggesting there was no press release issued for this study. Odd, that — especially if the decline is as real and significant as the authors suggest.
While no evidence was provided for a correlation of this decline in numbers to recent (2012-2016) sea ice decline, previous evidence from the region (Obbard et al. 2016) showed a decline in body condition was correlated only with much later than usual freeze-up, a situation that did not occur from 2012 to 2015 (freeze-up was late in the fall of 2016 but occurred months after the Obbard et al. (2018) survey was completed).
Moreover, the paper reports that a decline in survival of yearling cubs (from 12% of the population in 2011 to 5% in 2016) was not associated with especially poor sea ice conditions in spring. We are likely to see a follow-up paper next year reporting the body condition and sea ice data from this study (as for the previous survey: Obbard et al. 2015, 2016), but there is no suggestion in this paper that body condition declined further from 2011/2012 levels or that sea ice conditions deteriorated markedly enough after 2012 to precipitate a population decline.
Some background for this subpopulation is available in an earlier (2013) post here. Relative to Western Hudson Bay (WH), sea ice breaks up later in the summer in Southern Hudson Bay (SH) but freeze-up is also later in the fall: overall, however, the change in sea ice coverage since 1980 has been similar for both regions.
During breakup, sea ice usually remains late in the season along the southwest shore (Ontario) of Hudson Bay, so SH bears come ashore later than WH bears:
What the media reported
From the CTV News story (1 June 2018):
“An exhaustive survey of the world’s most southerly polar bears has found a significant drop in their numbers that scientists fear could be climate change finally taking its toll on a population whose health has long been in decline.
“If this trend is real and if it continues, I think we happened to have caught it just as it started to go over a cliff,” said Martyn Obbard, lead author of the paper that appeared this week in the journal Arctic Science.
Obbard looked at the southern Hudson Bay population, a group of polar bears that live on the southeastern shores of Hudson Bay and into James Bay. His co-authors include scientists from the governments of Ontario, Quebec and Nunavut, as well as the United States.
The study found that the number of those bears has declined about 17 per cent in the past five years, from 943 to 780.
Disturbingly, the number of yearlings also fell. Aerial surveyors said 12 per cent of the population were yearlings in 2011; that number has fallen to five per cent.
“Many adult females may still be producing litters, but they may be less successful in raising cubs,” says the paper.
A series of studies over the past few years has previously documented concerns about these bears.
They were getting smaller and skinnier as sea ice in the bays shrank, researchers found.
By 2012, the local bears were already spending on average 30 days longer on land than they did in 1980, depriving them of their critical, fat-rich diet of seals.
But the overall population of southern Hudson Bay bears was still considered stable. Obbard did the last survey of their numbers in 2011 and his count was in line with 25 years of previous work.”
What the study says about numbers
“The estimate from the current survey is not significantly different from the abundance estimate for the 2011/12 survey of 943 bears (95% CI 658–1350; Obbard et al. 2015) based on the confidence intervals overlapping the means. However, because polar bears are listed as Threatened in Ontario under the Endangered Species Act, we wished to better inform managers about the status of the subpopulation.
To do this, we used Monte Carlo simulations to explore the difference between the two estimates. We specified a lognormal distribution defined by the point estimates and SEs of the two surveys and randomly sampled from each of these distributions to represent potential population sizes for the two surveys (n = 1,000,000 iterations).
Sixty-one percent of iterations yielded a difference (i.e., a decline) of >100 between the estimates for 2011/12 and 2016 (Figs. 7a, b). In terms of proportional differences, 63% of the runs showed a decline of > 10% between 2011/12 and 2016, 43% showed > 20% decline and 32% showed > 25% decline.
We summed the results of the perpendicular transect analysis including the coastal zone with the estimates of bears on small islands, yielding 763 bears (SE: 125.1). We obtained an estimate of 796 bears (SE: 92.1) by adding the estimates from the coastal contour transects, the perpendicular transects excluding the coastal region, and small islands. Averaging these figures yielded an estimate of 780 bears (SE: 111.1; 95% CI: 590–1029) in SH during the 2016 ice-free season.
Our results suggest that abundance in the Southern Hudson Bay subpopulation declined by 17% between 2011/12 and 2016.
Though the point estimates obtained for the 2011/12 and 2016 surveys were not significantly different, point estimates of abundance are typically used to assess status (PBSG 2010b). The suggestion of a decline in abundance is supported by the Monte Carlo simulations where about one-third showed a greater than 25% decline between the two estimates.“ [my bold]
Monte Carlo simulations are indeed a standard statistical method of exploring the relationship between two estimates. However, why wasn’t the method used to determine if the statistically insignificant increase for polar bear numbers in the Svalbard region (Barents Sea) surveyed in 2004 and 2015 (Aars et al. 2017) could be a real increase that should be noted by managers? And why wasn’t it used to determine if the 2016 statistically insignificant decline recorded for Western Hudson Bay since 2012 (Dyck et al. 2017) stood a chance of being a real decline worth noting?
It seems odd to me that the Monte Carlo simulation technique was used in this case of statistical insignificance but not in other similar instances.The most recent example I could find of its use was in Taylor et al. (2005), who used Monte Carlo simulations to estimate “the risk of depletion for a number of harvest scenarios.”
What the study says about cub survival
“The proportion of cubs [of the year] in all observations was similar in the 2011 and 2016 surveys, but the proportion of yearlings dropped from 12% to 5% (sample size from the 2012 portion of the original survey is too small for comparison). We are not aware of any evidence of reproductive synchrony in polar bear populations that would infer large variation in size of the year class of cubs born. Therefore, the low proportion of yearlings in the 2016 survey suggests low survival of cubs born in 2015.
Whether this represents an ongoing trend, or was the result of unusual conditions in spring and summer of 2015 is undetermined. However, breakup in eastern Hudson Bay was about 2 weeks later in 2015 than in 2010-2014 (Andrews et al. 2018) suggesting that ice conditions should not have had a negative influence on cub survival in 2015. The proportion of yearlings in the 2016 SH survey was similar to that observed in the WH surveys of 2011 and 2016 (3%).
Because the proportion of cubs remained high in 2016, many adult females in SH are still producing litters, but they may be less successful in raising cubs to yearling age.” [my bold]
Breakup was not only relatively late in 2015, freeze-up was also relatively early. This leaves no correlation between timing of sea ice changes or length of sea ice coverage as an explanation for poor yearling cub survival in 2015 (see ice charts below) but what is unknown are the conditions out on Hudson Bay over the winter and early spring with respect to sea ice thickness, snow cover, and polar bear survival:
Breakup 2015, early August:
Freeze-up 2015, end November (this is relatively early for SH, while “late” would be mid-to-late December):
What the study says about SH sea ice decline
Not a lot, except that 2015 did not have unusually low ice cover and that ice cover over SH has declined by about 30 days since 1980. But that tells us nothing about conditions since the last population survey in 2012 (i.e. 2012-2016) and how those conditions might correlate with the slight population decline recorded by the authors.
See ice graphs below from Regehr et al. (2016) document changes in sea ice coverage between 1979 and 2014, and show the SH decline (lower left, -0.68 days per year) was not as great as WH (lower right, -0.86 days per year):
Below is the sea ice data (1980-2012) from the previous SH population assessment study that looked at body condition vs. sea ice in 2011/2012 (Obbard et al. 2016) : grey dots are freeze-up dates, open dots are breakup dates, and solid circles indicate length of the ice cover season. Those authors found that body condition correlated with freeze-up dates only, not duration of ice coverage or breakup dates (note a step-change in length of ice coverage (black dots) at about 1995 or 1998, similar to the step-change noted for WH by Castro de la Guardia 2017 discussed here):
The slight population decline recorded by the authors of this study (Obbard et al. 2018) might be an early manifestation of a larger decline to come or it could be due to the inherent variation that comes with population size estimates.
Statistically, the 17% decline these authors found was not significant, even though Monte Carlo simulations suggested the decline might be meaningful. However, the appropriateness of using of Monte Carlo techniques to bolster a statistically insignificant result is not especially compelling in this case, given that the same technique was not used in any other recent polar bear population survey with similar statistically insignificant declines or increases. Why now, not then?
Indeed, Environment Canada’s 2017 update (February 2018) lists the SH subpopulation as stable along with Foxe Basin (the government report upon which this Obbard et al. study is based was completed early last year and I assume it was available to the Polar Bear Technical Committee when making its latest status update, shown below — if anyone knows that is not the case, let me know):
Aars, J., Marques,T.A, Lone, K., Anderson, M., Wiig, Ø., Fløystad, I.M.B., Hagen, S.B. and Buckland, S.T. 2017. The number and distribution of polar bears in the western Barents Sea. Polar Research 36:1, 1374125. DOI:10.1080/17518369.2017.1374125
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. 2016 Aerial survey of the Western Hudson Bay polar bear subpopulation. Final report, Nunavut Department of Environment, Wildlife Research Section, Iglolik, NU. http://www.gov.nu.ca/environnement/information/wildlife-research-reports#polarbear
Lunn, N.J., Servanty, S., Regehr, E.V., Converse, S.J., Richardson, E. and Stirling, I. 2016. Demography of an apex predator at the edge of its range – impacts of changing sea ice on polar bears in Hudson Bay. Ecological Applications 26(5): 1302-1320. DOI: 10.1890/15-1256
Obbard, M.E., Stapleton, S., Middel, K.R., Thibault, I., Brodeur, V. and Jutras, C. 2015. Estimating the abundance of the Southern Hudson Bay polar bear subpopulation with aerial surveys. Polar Biology 38:1713-1725.
Obbard, M.E., Cattet, M.R.I., Howe, E.J., Middel, K.R., Newton, E.J., Kolenosky, G.B., Abraham, K.F. and Greenwood, C.J. 2016. Trends in body condition in polar bears (Ursus maritimus) from the Southern Hudson Bay subpopulation in relation to changes in sea ice. Arctic Science 2:15-32. http://www.nrcresearchpress.com/doi/abs/10.1139/AS-2015-0027#.VvFtlXpUq50
Obbard, M.E., Stapleton, S., Szor, G., Middel, K.R., Jutras, C. and Dyck, M. 2018. Estimating the abundance of the Southern Hudson Bay polar bear subpopulation with aerial surveys. Arctic Science https://doi.org/10.1139/AS-2018-0004
The Southern Hudson Bay polar bear (Ursus maritimus Phipps, 1774) subpopulation is considered stable but conflicting evidence lends uncertainty to that designation. Capture-recapture studies conducted 1984-86 and 2003-05 and an aerial survey conducted 2011/12 suggested abundance was likely unchanged since the mid-1980s. However, body condition and body size declined since then, and duration of sea ice decreased by about 30 days. Due to conflicting information on subpopulation status and ongoing changes in sea ice, we conducted another aerial survey in 2016 to determine whether abundance had changed. We collected data via mark-recapture distance sampling and double-observer protocols. Results suggest abundance declined 17% from 943 bears (95% CI 658-1350) in 2011/12 to 780 (95% CI 590–1029) in 2016. The proportion of yearlings declined from 12% of the population in 2011 to 5% in 2016, whereas the proportion of cubs remained similar (16% in 2011 vs. 19% in 2016) suggesting low survival of the 2015 cohort. In a warming Arctic, duration of sea ice is predicted to continue to decline in Hudson Bay affecting all ice-dependent wildlife; therefore, further monitoring of this subpopulation is warranted. We recommend a conservative approach to harvest management and repeating the aerial survey in 2021.
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#
Regehr, E.V., Laidre, K.L, Akçakaya, H.R., Amstrup, S.C., Atwood, T.C., Lunn, N.J., Obbard, M., Stern, H., Thiemann, G.W., & Wiig, Ø. 2016. Conservation status of polar bears (Ursus maritimus) in relation to projected sea-ice declines. Biology Letters 12: 20160556. http://rsbl.royalsocietypublishing.org/content/12/12/20160556 Supplementary data here.
Taylor, M.K., Laake, J., McLoughlin, P.D., Born, E.K., Cluff, H.D., Ferguson, S.H., Rosing-Asvid, A., Schweinsburg, R. and Messier, F. 2005. Demography and viability of a hunted population of polar bears. Arctic 58:203-214.
Previous papers/reports on this subpopulation
Obbard, M.E., Cattet, M.R.L., Moody, T., Walton, L.R., Potter, D., Inglis, J. and Chenier, C. 2006. Temporal trends in the body condition of southern Hudson Bay polar bears. Ontario Ministry of Natural Resources, Applied Research and Development Branch, Sault Ste, Marie, Canada. Climate Change Research Information Note 3. Available from http://sit.mnr.gov.on.ca
Obbard, M.E., McDonald, T.L., Howe, E.J., Regehr, E.V. and Richardson, E.S. 2007. Polar bear population status in southern Hudson Bay, Canada. Administrative Report, U.S. Department of the Interior- U.S. Geological Survey, Reston, VA.
Obbard, M.E., Middel, K.R., Stapleton, S., Thibault, I., Brodeur, V. and Jutras, C. 2013. Estimating abundance of the Southern Hudson Bay polar bears subpopulation using aerial surveys, 2011 and 2012. Ontario Ministry of Natural Resources, Wildlife Research and Monitoring Section, Science and Research Branch, Wildlife Research Series 2013-01. Peterborough, Ontario. Pdf here.
Obbard, M. E. and Walton, L.R. 2004. The importance of polar bear provincial park to the southern Hudson Bay polar bear population in the context of future climate change. Proceedings of the Parks Research Forum of Ontario (PRFO):105-116. [added July 26, 2013] pdf here.
June 4, 2018 at 01:24AM