Year of Award


Document Type


Degree Type

Master of Science (MS)

Degree Name

Wildlife Biology

Department or School/College

College of Forestry and Conservation

Committee Chair

Mark Hebblewhite

Commitee Members

Paul M. Lukacs, Michael S. Mitchell, Kelly M. Proffitt


Bayesian state-space models, Cox proportional hazards models, life-stage simulation analysis, time-to-event, ungulate, Yellowstone National Park


University of Montana

Subject Categories

Natural Resources and Conservation


The recolonization of gray wolves (Canis lupus) in multiple carnivore systems has been correlated with declining recruitment of juvenile ungulates, but the importance of wolves relative to other established carnivores is uncertain. The realized effect of carnivores may depend on compensation from forage and winter weather severity, which may mediate juvenile vulnerability to predation. Also, in high predation systems, the importance of juvenile recruitment may be reduced for ungulate population dynamics. We first tested for the effects of risk factors on annual elk (Cervus elaphus) calf survival and estimated cause-specific mortality rates. Secondly, we investigated the relative importance of different vital rates, including juvenile survival, on elk population dynamics using Bayesian integrated population modeling. Summer survival probability was less variable and averaged 0.55 (95% CI = 0.47, 0.63), while winter survival varied significantly across study years (P = 0.002) and averaged 0.73 (95% CI = 0.64, 0.81). Mountain lions (Felis concolor) dominated elk calf mortality in summer and winter, with estimated cause-specific mortality rates of 0.14 (95% CI = 0.09, 0.20) and 0.12 (95% CI = 0.07, 0.18), respectively, compared to only 0.01 (95% CI = 0.002, 0.05) and 0.03 (95% CI = 0.01, 0.07) caused by wolves, demonstrating that despite wolves recolonizing in this system, mountain lions were the leading cause of mortality. We found that elk calf survival increased in areas of high forage biomass in summer, and decreased strongly with increasing predation risk from mountain lions, but only in winter (Hazard Ratio = 2.84, 95% CI = 1.37, 5.88, P = 0.005). We found no evidence that forage availability or winter weather severity mediated vulnerability to mountain lion predation risk, indicating that the effects of mountain lion predation were relatively constant. Bayesian integrated population models revealed that on average adult female survival explained more of the variation in population growth rate than elk calf survival, while pregnancy rates were less important for population growth. By decomposing annual calf survival, we found that the summer and winter survival periods were nearly equivalent in importance for elk population dynamics in one population. The importance of carnivores for juvenile ungulate recruitment varies across ecological systems depending on relative carnivore densities, and mountain lions may be the most important carnivore for ungulates in more settings than anticipated. Given the high importance of adult female survival for elk population growth rates in systems with established carnivore populations, managers may need to reduce adult female harvest as carnivores recolonize, especially in areas with less productive habitat for elk.



© Copyright 2015 Daniel R. Eacker