Addressing the Impact of Time-Dependent Social Groupings on Animal Survival and Recapture Rates in Mark-Recapture Studies

Alexandru M. Draghici, The University of Western Ontario

Abstract

Mark-recapture (MR) models typically assume that individuals under study have independent survival and recapture outcomes. One such model of interest is known as the Cormack-Jolly-Seber (CJS) model. In this dissertation, we conduct three major research projects focused on studying the impact of violating the independence assumption in MR models along with presenting extensions which relax the independence assumption. In the first project, we conduct a simulation study to address the impact of failing to account for pair-bonded animals having correlated recapture and survival fates on the CJS model. We examined the impact of correlation on the likelihood ratio test (LRT), the 𝑐 ̂ correction, and the achieved coverage of 95% confidence intervals around the recapture and survival probabilities estimated from the CJS model. We find that correlated fates between mated animals may result in underestimated standard errors for parsimonious models, deflated LRT statistics, and underestimated values of 𝑐 ̂ for models taking sex-specific effects into account. In the second project, we present a novel conditional data approach to estimating recapture and survival correlations between mates. We provide a simulation study which demonstrates that for sufficiently large sample sizes the estimators of recapture and survival correlations between mated pairs are unbiased and achieve at least nominal coverage for 95% confidence intervals. The study shows that the variance correction using an alternative 𝑐 ̂ estimator addresses the issue of undercoverage and demonstrate the application of my model extension to a mark-recapture dataset of Harlequin ducks (Histrionicus histrionicus), a large monogamous waterfowl species. The final project in this work is focused on presenting extensions to both the CJS and Jolly-Seber (JS) model which allow mortality of members within a group to influence the future survival outcomes of remaining members with Bayesian methods. We conduct a simulation study which demonstrated that the models produce unbiased estimates and credible intervals which achieve nominal coverage. Finally, we apply the CJS model extension to a dataset of Wild Turkeys (Meleagris gallopavo silvestris) and find that there is evidence to suggest that mortality results in reduced survival rates for remaining group members.