Barents Sea polar bear status and sea ice declines

So far, I’ve not discussed the Barents Sea subpopulation in very much detail, except in comparison to other groups. For example, the Barents is considered to be the same type of sea ice “ecoregion” as the Chukchi Sea and the Southern Beaufort (discussed here). Previous studies on the Barents Sea polar bear population (Derocher 2005) indicate it may have recovered from extreme levels of overhunting (discussed here) and had stabilized, or was increasing very slowly, as early as 2002 (discussed here) — similar to what has happened in Davis Strait (discussed here).

Figure 1. Polar bear subpopulations, with the Barents Sea region highlighted; map courtesy the IUCN Polar Bear Specialist Group (PBSG), extra labels added.

The most recent Barents Sea population estimate was done in 2004 (2,650; range ~1900-3600), based on an aerial survey (Aars et al. 2009). Aerial surveys are the only practical method of establishing population counts in regions like this where many bears never set foot on land. The previous estimate for the Barents (1982) was “2,000-5,000” but its accuracy was considered “poor” (discussed here).

The IUCN Polar Bear Specialist Group (PBSG), in their most recent report, lists the Barents Sea population as “data deficient” for status, current trend and estimated risk of decline within 10 years (Obbard et al. 2010:62, Table 1) and the “notes” for this entry say:

“Population estimate is based on a new aerial survey. There was likely an increase in the subpopulation size after 1973 until recently. Current growth trend is unknown.”

This 2004 estimate is now almost a decade old and potentially no longer an accurate representation of what’s happening in the Barents Sea. The most up-to-date information has not yet been published but it is available online. It’s eye-opening to say the least, if only that it appears to be yet another example of a polar bear population that is so far not showing signs of being harmed by sea ice declines, as I’ve discussed before (here).

[Update October 15, 2013: I’ve simplified the text discussion and figure regarding the Aars and Andersen denning study from the original posted]

On the website of the Norwegian Environmental Monitoring of Svalbard and Jan Mayen (MOSJ), with data on polar bears submitted by Jon Aars and Magnus Andersen, they have the following introduction:

“The most important man-made threat factor to polar bears are loss of sea ice habitat due to climate warming. To be able to detect the effects of such stressors on an early stage, the population monitoring have been focused on reproduction and body condition”.

So, there are two issues: sea ice decline and the effects that sea ice decline might have on reproductive success and body condition.

Sea ice changes
The Barents Sea has experienced some declines in summer/fall sea ice over the last 30 years or so, although year-to-year variability is high. For example, Aars and Andersen provide graphs showing the number of days per year with >60% ice cover in late fall/early winter (October 1 to December 31) around two known maternity den areas for pregnant polar bear females: Hopen Island, in the far south, and Kongsoya Island, in the Kong Karl’s Land group east of Spitsbergen (Fig.2). The amount of ice available at this time of year determines whether or not females can get to the islands to make their dens. [only pregnant females routinely enter dens for the winter; others do sporadically, especially during storms or the darkest part of the season, see previous post here].

Figure 2. Map showing the location of the two islands (Kongsgoya and Hopen) where research on polar bear maternity den use vs. sea ice cover in fall (October 1 to December 31) has been reported by Aars and Andersen. Map from Wikipedia. Click to enlarge.

Aars and Andersen (here) say this about the fall sea ice declines:

“The number of days where there are more than 60 % sea ice cover has a strong influence on the number of females that den on Kongsøya; few days with sufficient ice cover leads to few females entering dens.

The number of days where there are more than 60 % sea ice cover has a strong influence on the number of females that den on Hopen; few days with sufficient ice cover leads to few females entering dens. Since 1979, when sea ice monitoring started, there has been a marked decrease in the number of days with sea ice cover around Hopen, and since the den counts started in the 1990s, there has been a similar decrease in the number of dens on the island.”

OK – less sea ice around these islands during the denning period in late fall causes females to den elsewhere. Hardly surprising and hardly a catastrophe: in the Barents Sea, bears are known to den on the first year drifting pack ice that covers the Barents Sea from winter through early summer (Mauritzen et al. 2001), which means denning on the pack ice is always an option if a female can’t get to land (even if it’s not what all females prefer).

[Jon Aars, in a recent published paper (Aars 2013), discusses how difficult it can be to find these maternity dens — timing is everything]

What about sea ice cover in the spring and early summer? A few months ago (in The Guardian, August 6, 2013), polar bear biologist Ian Stirling was quoted as saying:

“Most of the fjords and inter-island channels in Svalbard did not freeze normally last winter…”

That may be true. It is also true that sea ice area in the Barents Sea has declined, especially since 1998, which was a high ice year, as shown by the work of MOSJ researchers Olga Pavlova and Sebastian Gerland (Fig. 3) – 2006 was the lowest measured since 1979.

Figure 3. Barents Sea ice area in April, 1979-2012. From Pavlova and Gerland, MOSJ. Original caption: “The figure shows the average ice area in April in the Barents Sea, the month with the highest prevalence of ice in the area. Data are shown as monthly mean values for each year (thin blue curve), running mean values over 3 years (thick blue curve), and linear trend throughout the period (red line). The inter annual variation is large, but there is also an obvious negative trend for April through the period. The lowest record of extent was observed in April 2006.”

However, did any of these changes in ice coverage have a negative effect on polar bear body condition or their reproductive success? Let’s take a look at the data.

Body condition and reproduction
Aars and Andersen have posted their data on litter size, number of cubs per female (cubs of the year or COY), proportion of females with cubs (COYs and yearlings) and condition of adult males.

I’ve referred to their data on body condition of adult males in spring previously (here), which includes figures through spring of this year (1993-2013, March through May). Aars and Andersen found no significant trend over time in the condition of adult males in spring (Fig. 4, below).

Moreover, 2013 contained no “outliers” (bears in much poorer condition than others that year) and several years (e.g. 1998, 2000) had more bears that were in poor condition than were found in 2013. These results indicate that winter/spring ice conditions in 2013 did not result in an unusual number of bears in poor body condition compared to previous years, back to 1993 – contrary to what Stirling has suggested.

Figure 4. Body condition of adult males in Svalbard in spring, from Aars and Andersen. Original caption: “Body condition index of adult male polar bears caught in spring (Mar-May) in the period 1993-2013.” There is no significant trend over time, according to the authors. Black dots are outliers. The “body condition index” values (not defined) appear to be anomalies (i.e. “0” is “normal”) – I’ll correct this if I find out otherwise.

How about cub production? The researchers found a statistically significant decline in the number of new cubs (cubs-of-the-year, COY) per adult female over time (1993-2013), although the year-to-year variation was sometimes quite high (Fig. 5).

Figure 5. Number of cubs of the year per adult female, from Aars and Andersen, MOSJ. Original caption: “The figure shows the number of cubs of the year (COYs) pr adult female based on data from the annual capture-recapture program 1993-2013. The dotted line shows a significant decreasing trend over time (p = 0.049).” [my bold]

They also found a slight decrease (non-significant) in the proportion of females with COYs over time (not shown here), no change over time in the proportion of females with COYs (not shown here), and a non-significant trend of decreasing litter size over time” (Fig. 6). For litter sizes, the range of values from one year to the next is huge and the sample sizes vary similarly (from 2 to 25!) – no wonder there’s not a significant trend.

Figure 6. Average litter size, from Aars and Andersen, MOSJ. Original caption: “Average litter size (cubs of the year); data from the annual capture-recapture program 1993-2013. There is a non-significant trend of decreasing litter size over time (dotted line). The number of litters are given above each annual estimate.” [my bold and underline]

In summary, it appears from the latest data available that despite a decline in Barents Sea ice during the spring and fall, only one critical polar bear life history parameter out of five (number of COYs per adult female) showed any correlation with sea ice coverage. In other words, Barents Sea polar bears appear to be doing well, as measured by body condition and reproductive success, despite sea ice declines, which means it is doubtful if their population size has declined significantly since 2004.

References
Aars, J. 2013. Variation in detection probability of polar bear maternity dens. Polar Biology 36:1089-1096. http://link.springer.com/article/10.1007%2Fs00300-013-1331-7

Aars, J., Marques, T.A., Buckland, S.T., Andersen, M., Belikov, S., Boltunov, A., and Wiig, Ø. 2009. Estimating the Barents Sea polar bear subpopulation. Marine Mammal Science 25:35-52. http://onlinelibrary.wiley.com/doi/10.1111/j.1748-7692.2008.00228.x/abstract

Derocher 2005. Population ecology of polar bears at Svalbard, Norway. Population Ecology 47:267-275.
http://www.springerlink.com.ezproxy.library.uvic.ca/content/765147518rp35613/fulltext.pdf

Mauritzen, M., Derocher, A.E. and Wiig, Ø. 2001.  Space-use strategies of female polar bears in a dynamic sea ice habitat. Canadian Journal of Zoology 79:1704-1713. http://www.nrcresearchpress.com/doi/abs/10.1139/z01-126#.Ul1dxhDkSUk

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, IUCN.