4. Discussion

We examined intra-population variation in energy density and storage energy, temporal dynamics in energetics, and the influence of sea ice dynamics on WH polar bear population energetics from 1985 to 2018. All age/sex classes, except adult males and adult females with offspring declined in energy density over time, all classes except adult females with offspring declined in storage energy over time, and reduced energy values were associated with earlier sea ice breakup across classes. Furthermore, we found that the WH population energy density and storage energy both declined significantly over time in relation to earlier sea ice breakup and longer lagged open water periods. This research is important for monitoring trends in individual physiological condition, understanding implications for population dynamics, and predicting future responses of polar bears to climate warming.

Age/sex class patterns

We found variation in energetic dynamics from 1985 to 2018 among WH polar bear age/sex classes. All classes except adult males and adult females with offspring showed decreases in energy density over time, and all classes except adult females with offspring declined in storage energy over time, with the most significant declines for solitary adult females and yearlings. Energy density is determined as the ratio of storage energy to body mass (Molnár et al. 2009) and was less variable than storage energy. In contrast, storage energy indicates the total amount of energy in an individual (Molnár et al. 2009) and is therefore more sensitive to changes in body condition given that the amount of storage energy is lower at smaller body sizes. Decreases in storage energy indicate that the bears had less energy available for maintenance, growth, and survival (Molnár et al. 2009; Sciullo et al. 2016). These results are similar to Sciullo et al. (2016) where WH polar bear storage energy declined from 2004-2013 across classes. Due to the relationship between energy reserves/body condition and fitness (Jakob et al. 1996; Sciullo et al. 2016), the observed reductions in available energy will influence survival and reproduction, with consequences for individual fitness. The significant reduction in energy density and storage energy over time for solitary adult females and the significant reduction in storage energy for yearlings further indicates the vulnerability of these classes to future environmental changes. The small body size, dietary constraints, energetic demands of growth, and inexperienced hunting skills of younger bears make them more vulnerable to reductions in sea ice and thus prey availability (Rode et al. 2010; Thiemann et al. 2011a; Pilfold et al. 2016; Johnson et al. 2019; Laidre et al. 2020). In contrast, adult males can best buffer against sub-optimal conditions given their larger body size, broader diets, more effective hunting skills, and ability to take kills away from smaller bears (Stirling 1974; Regehr et al. 2007; Thiemann et al. 2011a; Pilfold et al. 2016; Johnson et al. 2019). These patterns highlight the importance of continued monitoring of the condition of young bears.
The reproductive status of adult female polar bears in WH influenced their energy patterns. Solitary adult females had higher energy density and storage energy than adult females with offspring, but solitary females experienced significant declines in both energy metrics over time whereas females with offspring had lower but relatively stable energy values. These results are consistent with observations that solitary adult females have higher body condition due to their accumulation of body fat in preparation for the energetic requirements of gestation and lactation (Atkinson & Ramsay 1995; Thiemann et al. 2006; Sciullo et al. 2016). The amount of energy a solitary adult female accumulates before denning determines the likelihood of successfully producing cubs as well as subsequent cub survival (Atkinson & Ramsay 1995; Derocher & Stirling 1994, 1996, 1998) and litter size (Laidre et al. 2020). Decreases in solitary adult female condition can therefore translate into a decline in cub survival and reproductive success, both of which have already been observed in WH (Derocher & Stirling 1995; Stirling et al. 1999). The observed declines in solitary adult female energy may reflect difficulty in accumulating sufficient resources. In contrast, females with offspring have lower energy reserves due to lactational energetic demands (Derocher et al. 1993; Arnould & Ramsay 1994; Atkinson & Ramsay 1995). There is likely a threshold of energetic reserves that is required to successfully reproduce (Molnár et al. 2010; Reimer et al. 2019). For instance, Derocher et al. (1992) found that the lowest weight of an adult female that was known to have successfully reproduced was 189 kg, while Robbins et al. (2012) indicated that females require 20% body fat when entering a den to successfully produce cubs. Similarly, our results indicated that adult females with offspring had relatively stable energy density (median: 19.8 MJ kg-1; Fig. 2) and storage energy (median: 2241 MJ; Fig. 3), suggesting energetic thresholds for reproduction. In agreement with Robbins et al. (2012), our results highlight the vulnerability of females with offspring due to their already-reduced energetic reserves, as well as the sensitivity of solitary adult females that need to accumulate sufficient energy in preparation for future reproductive events yet are nutritionally stressed.
Our study also demonstrated the association between age/sex class energetic patterns and environmental conditions. Reduced energy density and storage energy were associated with earlier sea ice breakup and this relationship was most significant for adult males, subadult males/females, and COY. These results are consistent with the relationship between earlier breakup and reduced body condition in WH (Stirling et al. 1999; Sciullo et al. 2016). Our finding that the lagged open water period was an important predictor for solitary adult female storage energy suggests that the previous year’s sea ice conditions influenced the ability of solitary females to accumulate energy reserves in preparation for reproduction. Similarly, Derocher & Stirling (1994) found that an adult female’s condition in the previous year was a strong determining factor for reproductive success in WH. In addition, the previous year’s breakup date influenced body condition of Davis Strait polar bears, whereby earlier breakup in the previous year forced bears ashore earlier, resulting in poorer body condition (Galicia et al. 2019), while Baffin Bay polar bear body condition was associated with the previous ice-free period (Laidre et al. 2020). Our observed decline in solitary adult female energy and the relationship with the lagged open water period suggests that females may not be able to recover from poor conditions in previous years, which may build up over time and affect lifetime reproductive success. Overall, our results indicate that polar bear energetic balances are negatively affected by sea ice declines and that vulnerable demographic groups include younger bears and adult females.

Population trends

WH total population energy density and storage energy declined significantly over the 34 year study period. Both energy metrics were correlated with estimated abundance, which indicates that the population energy decline was related to reduced population abundance. The WH population has declined from approximately 1185 to 806 bears over the period of 1987 to 2011 (Lunn et al. 2016) with a continued decline in recent years (ECCC unpublished data, 2019); furthermore, WH body condition has also declined over time (Derocher & Stirling 1995; Stirling et al. 1999; Sciullo et al. 2016). Reduced population abundance in addition to declining body condition of individuals may both contribute to the observed decline in the total energy stored in this population. Declines in individual energy balances and subsequent consequences for survival and reproduction illustrate the mechanism linking climate change and population dynamics (Yodzis & Innes 1992; Humphries et al. 2004; Molnár et al. 2009, 2010; Pagano et al. 2018). Understanding the ecological mechanisms behind demographic change is important for wildlife management and can improve our predictions about how populations may respond to future climate warming (Cherry et al. 2009; Pagano et al. 2018, Reimer et al. 2019).
We found that western Hudson Bay experienced significant long-term change in sea ice dynamics, with a lengthening of the open water period by approximately 9.9 days per decade. WH polar bear population energy density and storage energy were both significantly reduced when sea ice breakup was earlier and the lagged open water period was longer, demonstrating a linkage between declining sea ice and reduced energetic balances. Sea ice is probably the most important single mechanism influencing polar bear demographic responses in the changing Arctic marine ecosystem. Our results are consistent with the association between earlier breakup/later freeze-up and declining body condition (Stirling et al. 1999; Obbard et al. 2016; Sciullo et al. 2016; Laidre et al. 2020), altered foraging ecology (McKinney et al. 2009; Johnson et al. 2019), and reduced reproduction/survival rates and abundance (Regehr et al. 2007; Rode et al. 2010; Lunn et al. 2016; Obbard et al. 2018) in various polar bear populations including WH, Southern Beaufort Sea, Southern Hudson Bay, and Baffin Bay. Changes to energetic intake and expenditure in response to sea ice dynamics have consequences for energetic balances (Pagano et al. 2018). Polar bear energetic intake is reduced when breakup occurs earlier and freeze-up occurs later because the spring hunting period is shortened and bears are forced to fast on land for longer periods in poorer condition (Cherry et al. 2009, 2013; Rode et al. 2014, 2018). Meanwhile, energetic expenditure increases due to declines in optimal habitat (Durner et al. 2009; Stern & Laidre 2016), increasingly fragmentated and drifting sea ice (Mauritzen et al. 2003; Sahanatien & Derocher 2012; Auger-Méthé et al. 2016; Durner et al. 2017), and long-distance swims as a result of more open water (Durner et al. 2011; Pagano et al. 2012; Pilfold et al. 2017). We found that the open water period increased from 105 days in 1985 to 145 days in 2018, with a maximum of 166 days in 2015. An increase in the fasting period from 120 days in the 1980s to 165 days was predicted to lead to higher starvation rates for adult male polar bears in WH (Robbins et al. 2012), while 180 day fasting was predicted to lead to additional increases in starvation and mortality rates (Molnár et al. 2010, 2014; Pilfold et al. 2016). Similarly, our predictions indicated that at 180 day previous fasting period, population energy density and storage energy would decline to 58% and 63% lower than the mean estimated values, respectively. Decreases in the length of the spring foraging period are predicted to lead to declines in female polar bear expected fitness (Reimer et al. 2019) and higher fasting rates have occurred concurrently with reductions in survival and abundance (Cherry et al. 2009; Rode et al. 2014, 2018). Our predicted declines in WH population energy at longer fasting periods would have implications for individual fitness, population vital rates, and productivity. Moreover, the importance of the lagged open water period suggests that there are cumulative effects of prior conditions that affect the ability of bears to recover from previous nutritional stressors. Hudson Bay is expected to undergo continued sea ice loss in the future and WH polar bears are therefore at risk of further declines to energetic balances leading to reduced survival rates for young bears and decreased reproductive success, which may ultimately result in a functionally extinct population (Castro de la Guardia et al. 2013; Pilfold et al. 2016).

Ecosystem implications

While the Arctic marine ecosystem has already experienced various alterations due to climate warming (Wassman et al. 2011), our observed decline in population energy of a top predator may have further implications for ecosystem dynamics. Altered top predator population dynamics have the potential to cascade through ecosystems and influence trophic interactions and food web dynamics (Pace et al. 1999; Schmitz et al. 2000; Frank et al. 2005). For example, reduced body size of top predators has been associated with a weakening of predation pressure on lower trophic levels (Shackell et al. 2010). A potential consequence of reduced WH polar bear energetic balances is therefore altered trophic interactions with their primary prey species, ringed seals (Pusa hispida ). However, Hudson Bay ringed seals have similarly shown population declines over time (Young et al. 2015; Ferguson et al. 2017); thus, our limited understanding of changing predator-prey interactions in the Arctic would benefit from long-term monitoring of ecological parameters across multiple trophic levels. Our study provides insights into potential mechanisms linking WH polar bear population dynamics and sea ice decline. As the Arctic continues to warm, polar bears can act as an indicator species to improve our understanding of changing ecosystem dynamics (Rode et al. 2018). Our research reinforces the importance of long-term monitoring of individual physiological condition and broad population patterns.