The ecology of sleep in free-living animals

Abstract

Sleep is a behaviour that performs important physiological and cognitive functions including boosting the immune system, consolidating memories, and cleansing the brain of detrimental by-products of metabolism. Whilst lack of sufficient sleep impairs physiological and cognitive functions, we know little about how the requirement for sleep is managed by wild animals experiencing changing environmental conditions and facing the challenge of managing their time among competing fitness enhancing activities. In this thesis, I used cutting-edge and minimally-invasive biologgers to estimate sleep quantity (total sleep time/day), efficiency (number of sleep bouts/day, i.e. sleep fragmentation), and quality (duration of the longest sleep bout in a day) in free-living animals, and addressed long-standing questions on sleep ecology in the wild.

Exploiting a long-term dataset of acceleration data in a wild population of wild boar (Sus scrofa) in the Czech Republic, I investigated how sleep changes over the annual cycle in 28 individuals, in my first chapter. I developed a robust classification of sleep applied to the acceleration data, drawing on the peer-reviewed data on sleep in domestic pigs, and paired this with advanced Bayesian modelling that allows the estimation of consistent individual differences. In support of the hypothesis that the thermal environment is key for sleep, my analysis revealed that high ambient temperatures reduce sleep quantity, efficiency and quality, while snowfall leads to more consolidated and higher quality sleep. By explicitly considering inter- and intra-individual differences, my models also identified differences across individuals in sleep quantity and efficiency that were stable over time and suggest the presence of sleep phenotypes.

In my second chapter, I tested the hypothesis that sleep site characteristics contribute to explaining individual differences in sleep in wild boar, and also investigated the effect of landscape features on sleep quality (forest, farmland, water sources, and sources of anthropogenic disturbance). I found that the selection of sleep sites did not differ among individuals, and that individual differences in sleep quality are unrelated to sleep site selection. Further, among landscape features only increasing distance outside the forest affected sleep quality, where boar sleeping further outside forest had greater sleep quality. This likely reflects “anthropulse” effects observed during the COVID-19 pandemic, where increased human activity in the forest caused some boar to be displaced, with those who remained in the forest sleeping less.

In the third chapter I turned to a different study system, fallow deer (Dama dama) in an urban park (Phoenix Park, Dublin), and investigated how the development of sleep (sleep ontogeny) is shaped by the environment in free-living neonates. I followed a similar approach to the one used for wild boar to identify and quantify sleep in 19 males over the first 40 days of life. I found that sleep time in fawns is reduced and consolidated into fewer bouts as they age, and that larger fawns slept longer. Specifically consistent individual differences in sleep quantity, quality and efficiency, or sleep phenotypes, are present from birth, and correlate with the rate of development of sleep in the first 40 days of life such that shorter-sleeping individuals develop more rapidly. Contrary to wild boar, however, higher ambient temperature favoured longer sleep durations in fawns, likely because their small size and low fat reserves enable greater tolerance of high temperatures or through the use of sleeping sites as thermal refuges.

In my final chapter, I investigated factors that may explain why fawns would show individual differences in their sleep ontogeny. Stress is known to disrupt sleep quality and quantity, so I expected that fawns with higher cortisol to have reduced and less efficient sleep. In chapter 3 I found that heavier fawns slept longer, and so here I tested whether birth timing and maternal age, also linked to offspring survival and quality, were related to increased duration, reduced fragmentation, and greater quality of sleep. Contrary to predictions, fawns born later in the season slept longer, whilst maternal age and cortisol did not influence sleep. Birth timing is associated with important differences in the environment fawns are born into (time for growth until winter, predation risk). By sleeping less earlier in the season prior to the birth peak, fawns may reduce their risk of predation, and given that growth hormone is primarily produced during sleep, the requirements for faster growth to reach a greater pre-winter body mass may lead later born fawns to sleep more.

My research demonstrates that minimally-invasive, high-resolution biologgers offer a reliable tool to address fundamental questions on the ecology of sleep in wild animals over long time periods, and unveil how and why sleep varies between individuals. I have shown that whilst sleep is influenced by thermal environment, most of the variation in sleep quantity, efficiency, and quality is explained by individual-level effects. These results contrast with the view in sleep science that optimal sleep time is fixed and varies little among individuals and over the annual cycle. Importantly, my findings open new areas for research to unravel the causes of individual differences in sleep and the implications these have for the health and fitness of individuals.

Date of Award	Dec 2023
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Northern Ireland Department for the Economy
Supervisor	Isabella Capellini (Supervisor), Domhnall Jennings (Supervisor) & Luca Börger (Supervisor)