Assessing Calf Survival and the Quantitative Impact of Reproductive Success on the Declining Moose (Alces alces) Population in Northeastern Minnesota

Abstract

Ungulate reproductive success is an important driver of population performance. Long-term, the northeastern Minnesota moose (Alces alces) population declined 58% since 2006, yet aerial survey estimates indicate stability during 2012–2017. Because causal mechanisms for this decline were largely unknown, I investigated calf production, survival and cause-specific mortality of calves of global positioning system (GPS)-collared dams of this population. Global positioning system collars have been deployed on adult moose and other ungulates to study various aspects of their ecology, but until the current study they had not been fitted to moose neonates. In 2013 and 2014, I GPS-collared 54 neonates and monitored them for survival. In 2015, I monitored 50 calving dams for signs of neonatal mortality using behavioral cues, and assessed seasonal calf survival with aerial surveys. In 2013 and 2014, collared calf survival to 9 months was 0.34 (95% CI = 0.23–0.52), whereas uncollared calf survival in 2015 was estimated at 0.40 (95% CI = 0.30–0.54). Survival in all 3 years dropped dramatically in the first 50 days of life; hazard peaked at about 15 days of age when dams and calves departed calving sites. Predation was the leading cause of death of collared calves (84% of mortalities); wolf (Canis lupus) predation accounted for 77% of deaths. The cumulative probability of succumbing to all forms of mortality by 9 months of age was 69.8%. I documented higher wolf predation than other recent studies with similar predator communities. Habitat use during calving and the energetically demanding post-parturition period can be an important determinant of neonatal survival. I surveyed and compared habitat characteristics of pre-calving, calving, peak-lactation, and mortality sites at a fine and broad scale. I also compared mortality sites of calves killed by wolves versus other causes, as well as calving sites of cows that successfully reared a calf to winter to those that did not. Cows tended to move to areas of more conifer forest cover to calve. During peak-lactation, cows and their calves used steeper areas with abundant forage, high concealment, and less conifer cover. Mortality sites were more level than other site types. Generally, wolf-kill sites were farther from roads. Cows that successfully reared a calf to winter typically calved in areas with more deciduous forest and less forested wetland cover than cows whose calves died. I estimated population growth rate (λ) using adult survival and calf recruitment data from the Minnesota Department of Natural Resources demographic studies and the Recruitment-Mortality (R-M) Equation, and compared this estimate to those calculated using data from the annual aerial survey. Measures of recruitment were similar whether derived from the survey or collaring studies, but generally were higher in the collared sample. More recent calculations of λ (derived from the survey [2012–2017 stabilization; 1.02 ± 0.16 (SE)] and using the R-M Equation [1.04 ± 0.04]) indicated growth over the next 50 years. However, the stochastic model using parameters from the adult and calf studies indicated an average decline in the population over the next 50 years (λ = 0.94 ± 0.004). Habitat improvement projects for moose should consider not only forage requirements, but also cover, slope, and road density. Identifying specific causes of calf mortality and understanding their relations to various landscape characteristics and other extrinsic factors should yield insight into mechanisms contributing to the declining moose population in northeastern Minnesota and serve as a basis for ecologically sound management responses.