Demographic processes and behaviour of snapping turtles (Chelydra serpentina) in the context of past catastrophes and ongoing threats

Abstract

Lifetime patterns of somatic growth, reproduction, and survival comprise life history, which links individual traits to the vital rates that determine the properties of populations, such as generation time, potential rate of increase, and responses to environmental perturbation. Individual lifehistory traits, such as survival, age at first reproduction, reproductive frequency, and the size and number of offspring covary along a limited number of dimensions forming the pace-of-life continuum because they are tightly linked by trade-offs and constraints. Furthermore, variation in life history also covaries with morphological, physiological, and behavioural traits. This dissertation focuses on interconnectedness of life-history traits with social behaviour, population dynamics, and conservation. The Algonquin long-term field study of Snapping Turtles (Chelydra serpentina) provides a unique opportunity to analyze these relationships in a longlived organism with a slow life history by building upon a productive foundation of previous research. Turtles‘ slow life history, low and variable juvenile recruitment, and reliance on high adult survivorship makes them vulnerable to anthropogenic threats resulting in turtles being disproportionately imperilled. In Chapter 1, I analyzed the patterns of abundance and survival during and after a population catastrophe and revealed individuals transitioning between sites in a connected population but no recovery over 23 years. Because of their cryptic behaviour, the mating system of Snapping Turtles was poorly known, so in Chapter 2 I quantify sexual size dimorphism and frequency of wounds to infer patterns of intraspecific aggression consistent with a mating system mediated by male combat. The third chapter focused on the somatic growth component of life-history by refining growth modelling by developing a model of seasonal variation in growth rates. In Chapter 4, I examine the demography of Snapping Turtles dispersing across roads by testing hypotheses based on the mating system revealed in Chapter 2 using a demographic model parameterized with survivorship estimated in Chapter 1 and the growth modeling approach developed in Chapter 3. I show that juveniles are overrepresented on roads and face higher mortality risk and that the lost reproductive value of juveniles killed on roads contributes substantially to the overall burden of road mortality in this long-lived species.